Merge branch 'for-4.13/block' of git://git.kernel.dk/linux-block

Pull core block/IO updates from Jens Axboe:
 "This is the main pull request for the block layer for 4.13. Not a huge
  round in terms of features, but there's a lot of churn related to some
  core cleanups.

  Note this depends on the UUID tree pull request, that Christoph
  already sent out.

  This pull request contains:

   - A series from Christoph, unifying the error/stats codes in the
     block layer. We now use blk_status_t everywhere, instead of using
     different schemes for different places.

   - Also from Christoph, some cleanups around request allocation and IO
     scheduler interactions in blk-mq.

   - And yet another series from Christoph, cleaning up how we handle
     and do bounce buffering in the block layer.

   - A blk-mq debugfs series from Bart, further improving on the support
     we have for exporting internal information to aid debugging IO
     hangs or stalls.

   - Also from Bart, a series that cleans up the request initialization
     differences across types of devices.

   - A series from Goldwyn Rodrigues, allowing the block layer to return
     failure if we will block and the user asked for non-blocking.

   - Patch from Hannes for supporting setting loop devices block size to
     that of the underlying device.

   - Two series of patches from Javier, fixing various issues with
     lightnvm, particular around pblk.

   - A series from me, adding support for write hints. This comes with
     NVMe support as well, so applications can help guide data placement
     on flash to improve performance, latencies, and write
     amplification.

   - A series from Ming, improving and hardening blk-mq support for
     stopping/starting and quiescing hardware queues.

   - Two pull requests for NVMe updates. Nothing major on the feature
     side, but lots of cleanups and bug fixes. From the usual crew.

   - A series from Neil Brown, greatly improving the bio rescue set
     support. Most notably, this kills the bio rescue work queues, if we
     don't really need them.

   - Lots of other little bug fixes that are all over the place"

* 'for-4.13/block' of git://git.kernel.dk/linux-block: (217 commits)
  lightnvm: pblk: set line bitmap check under debug
  lightnvm: pblk: verify that cache read is still valid
  lightnvm: pblk: add initialization check
  lightnvm: pblk: remove target using async. I/Os
  lightnvm: pblk: use vmalloc for GC data buffer
  lightnvm: pblk: use right metadata buffer for recovery
  lightnvm: pblk: schedule if data is not ready
  lightnvm: pblk: remove unused return variable
  lightnvm: pblk: fix double-free on pblk init
  lightnvm: pblk: fix bad le64 assignations
  nvme: Makefile: remove dead build rule
  blk-mq: map all HWQ also in hyperthreaded system
  nvmet-rdma: register ib_client to not deadlock in device removal
  nvme_fc: fix error recovery on link down.
  nvmet_fc: fix crashes on bad opcodes
  nvme_fc: Fix crash when nvme controller connection fails.
  nvme_fc: replace ioabort msleep loop with completion
  nvme_fc: fix double calls to nvme_cleanup_cmd()
  nvme-fabrics: verify that a controller returns the correct NQN
  nvme: simplify nvme_dev_attrs_are_visible
  ...

1 files changed, 66 insertions, 22 deletions

diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
index ab7cbb144f3f..2e6a5361baf0 100644
--- a/drivers/lightnvm/pblk-rl.c
+++ b/drivers/lightnvm/pblk-rl.c
@@ -23,11 +23,35 @@ static void pblk_rl_kick_u_timer(struct pblk_rl *rl)
 	mod_timer(&rl->u_timer, jiffies + msecs_to_jiffies(5000));
 }
 
+int pblk_rl_is_limit(struct pblk_rl *rl)
+{
+	int rb_space;
+
+	rb_space = atomic_read(&rl->rb_space);
+
+	return (rb_space == 0);
+}
+
 int pblk_rl_user_may_insert(struct pblk_rl *rl, int nr_entries)
 {
 	int rb_user_cnt = atomic_read(&rl->rb_user_cnt);
+	int rb_space = atomic_read(&rl->rb_space);
 
-	return (!(rb_user_cnt + nr_entries > rl->rb_user_max));
+	if (unlikely(rb_space >= 0) && (rb_space - nr_entries < 0))
+		return NVM_IO_ERR;
+
+	if (rb_user_cnt >= rl->rb_user_max)
+		return NVM_IO_REQUEUE;
+
+	return NVM_IO_OK;
+}
+
+void pblk_rl_inserted(struct pblk_rl *rl, int nr_entries)
+{
+	int rb_space = atomic_read(&rl->rb_space);
+
+	if (unlikely(rb_space >= 0))
+		atomic_sub(nr_entries, &rl->rb_space);
 }
 
 int pblk_rl_gc_may_insert(struct pblk_rl *rl, int nr_entries)
@@ -37,7 +61,7 @@ int pblk_rl_gc_may_insert(struct pblk_rl *rl, int nr_entries)
 
 	/* If there is no user I/O let GC take over space on the write buffer */
 	rb_user_active = READ_ONCE(rl->rb_user_active);
-	return (!(rb_gc_cnt + nr_entries > rl->rb_gc_max && rb_user_active));
+	return (!(rb_gc_cnt >= rl->rb_gc_max && rb_user_active));
 }
 
 void pblk_rl_user_in(struct pblk_rl *rl, int nr_entries)
@@ -77,33 +101,32 @@ static int pblk_rl_update_rates(struct pblk_rl *rl, unsigned long max)
 	unsigned long free_blocks = pblk_rl_nr_free_blks(rl);
 
 	if (free_blocks >= rl->high) {
-		rl->rb_user_max = max - rl->rb_gc_rsv;
-		rl->rb_gc_max = rl->rb_gc_rsv;
+		rl->rb_user_max = max;
+		rl->rb_gc_max = 0;
 		rl->rb_state = PBLK_RL_HIGH;
 	} else if (free_blocks < rl->high) {
 		int shift = rl->high_pw - rl->rb_windows_pw;
 		int user_windows = free_blocks >> shift;
 		int user_max = user_windows << PBLK_MAX_REQ_ADDRS_PW;
-		int gc_max;
 
 		rl->rb_user_max = user_max;
-		gc_max = max - rl->rb_user_max;
-		rl->rb_gc_max = max(gc_max, rl->rb_gc_rsv);
+		rl->rb_gc_max = max - user_max;
 
-		if (free_blocks > rl->low)
-			rl->rb_state = PBLK_RL_MID;
-		else
-			rl->rb_state = PBLK_RL_LOW;
+		if (free_blocks <= rl->rsv_blocks) {
+			rl->rb_user_max = 0;
+			rl->rb_gc_max = max;
+		}
+
+		/* In the worst case, we will need to GC lines in the low list
+		 * (high valid sector count). If there are lines to GC on high
+		 * or mid lists, these will be prioritized
+		 */
+		rl->rb_state = PBLK_RL_LOW;
 	}
 
 	return rl->rb_state;
 }
 
-void pblk_rl_set_gc_rsc(struct pblk_rl *rl, int rsv)
-{
-	rl->rb_gc_rsv = rl->rb_gc_max = rsv;
-}
-
 void pblk_rl_free_lines_inc(struct pblk_rl *rl, struct pblk_line *line)
 {
 	struct pblk *pblk = container_of(rl, struct pblk, rl);
@@ -122,11 +145,15 @@ void pblk_rl_free_lines_inc(struct pblk_rl *rl, struct pblk_line *line)
 
 void pblk_rl_free_lines_dec(struct pblk_rl *rl, struct pblk_line *line)
 {
-	struct pblk *pblk = container_of(rl, struct pblk, rl);
 	int blk_in_line = atomic_read(&line->blk_in_line);
-	int ret;
 
 	atomic_sub(blk_in_line, &rl->free_blocks);
+}
+
+void pblk_gc_should_kick(struct pblk *pblk)
+{
+	struct pblk_rl *rl = &pblk->rl;
+	int ret;
 
 	/* Rates will not change that often - no need to lock update */
 	ret = pblk_rl_update_rates(rl, rl->rb_budget);
@@ -136,11 +163,16 @@ void pblk_rl_free_lines_dec(struct pblk_rl *rl, struct pblk_line *line)
 		pblk_gc_should_stop(pblk);
 }
 
-int pblk_rl_gc_thrs(struct pblk_rl *rl)
+int pblk_rl_high_thrs(struct pblk_rl *rl)
 {
 	return rl->high;
 }
 
+int pblk_rl_low_thrs(struct pblk_rl *rl)
+{
+	return rl->low;
+}
+
 int pblk_rl_sysfs_rate_show(struct pblk_rl *rl)
 {
 	return rl->rb_user_max;
@@ -161,24 +193,36 @@ void pblk_rl_free(struct pblk_rl *rl)
 
 void pblk_rl_init(struct pblk_rl *rl, int budget)
 {
+	struct pblk *pblk = container_of(rl, struct pblk, rl);
+	struct pblk_line_meta *lm = &pblk->lm;
+	int min_blocks = lm->blk_per_line * PBLK_GC_RSV_LINE;
 	unsigned int rb_windows;
 
 	rl->high = rl->total_blocks / PBLK_USER_HIGH_THRS;
-	rl->low = rl->total_blocks / PBLK_USER_LOW_THRS;
 	rl->high_pw = get_count_order(rl->high);
 
+	rl->low = rl->total_blocks / PBLK_USER_LOW_THRS;
+	if (rl->low < min_blocks)
+		rl->low = min_blocks;
+
+	rl->rsv_blocks = min_blocks;
+
 	/* This will always be a power-of-2 */
 	rb_windows = budget / PBLK_MAX_REQ_ADDRS;
-	rl->rb_windows_pw = get_count_order(rb_windows) + 1;
+	rl->rb_windows_pw = get_count_order(rb_windows);
 
 	/* To start with, all buffer is available to user I/O writers */
 	rl->rb_budget = budget;
 	rl->rb_user_max = budget;
-	atomic_set(&rl->rb_user_cnt, 0);
 	rl->rb_gc_max = 0;
 	rl->rb_state = PBLK_RL_HIGH;
+
+	atomic_set(&rl->rb_user_cnt, 0);
 	atomic_set(&rl->rb_gc_cnt, 0);
+	atomic_set(&rl->rb_space, -1);
 
 	setup_timer(&rl->u_timer, pblk_rl_u_timer, (unsigned long)rl);
+
 	rl->rb_user_active = 0;
+	rl->rb_gc_active = 0;
 }