summaryrefslogtreecommitdiff
path: root/drivers/block
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2020-10-13 13:04:41 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2020-10-13 13:04:41 -0700
commit7cd4ecd9177b94af783b8e21de7c65b41a871342 (patch)
tree3ca393f3eaeeaad56d4ab60f87e28d7197b0ba21 /drivers/block
parent79ec6d9cac46d59db9b006bc9cde2811ef365292 (diff)
parent79cd16681acccffcf5521f6e3d8c7c50aaffca0a (diff)
Merge tag 'drivers-5.10-2020-10-12' of git://git.kernel.dk/linux-block
Pull block driver updates from Jens Axboe: "Here are the driver updates for 5.10. A few SCSI updates in here too, in coordination with Martin as they depend on core block changes for the shared tag bitmap. This contains: - NVMe pull requests via Christoph: - fix keep alive timer modification (Amit Engel) - order the PCI ID list more sensibly (Andy Shevchenko) - cleanup the open by controller helper (Chaitanya Kulkarni) - use an xarray for the CSE log lookup (Chaitanya Kulkarni) - support ZNS in nvmet passthrough mode (Chaitanya Kulkarni) - fix nvme_ns_report_zones (Christoph Hellwig) - add a sanity check to nvmet-fc (James Smart) - fix interrupt allocation when too many polled queues are specified (Jeffle Xu) - small nvmet-tcp optimization (Mark Wunderlich) - fix a controller refcount leak on init failure (Chaitanya Kulkarni) - misc cleanups (Chaitanya Kulkarni) - major refactoring of the scanning code (Christoph Hellwig) - MD updates via Song: - Bug fixes in bitmap code, from Zhao Heming - Fix a work queue check, from Guoqing Jiang - Fix raid5 oops with reshape, from Song Liu - Clean up unused code, from Jason Yan - Discard improvements, from Xiao Ni - raid5/6 page offset support, from Yufen Yu - Shared tag bitmap for SCSI/hisi_sas/null_blk (John, Kashyap, Hannes) - null_blk open/active zone limit support (Niklas) - Set of bcache updates (Coly, Dongsheng, Qinglang)" * tag 'drivers-5.10-2020-10-12' of git://git.kernel.dk/linux-block: (78 commits) md/raid5: fix oops during stripe resizing md/bitmap: fix memory leak of temporary bitmap md: fix the checking of wrong work queue md/bitmap: md_bitmap_get_counter returns wrong blocks md/bitmap: md_bitmap_read_sb uses wrong bitmap blocks md/raid0: remove unused function is_io_in_chunk_boundary() nvme-core: remove extra condition for vwc nvme-core: remove extra variable nvme: remove nvme_identify_ns_list nvme: refactor nvme_validate_ns nvme: move nvme_validate_ns nvme: query namespace identifiers before adding the namespace nvme: revalidate zone bitmaps in nvme_update_ns_info nvme: remove nvme_update_formats nvme: update the known admin effects nvme: set the queue limits in nvme_update_ns_info nvme: remove the 0 lba_shift check in nvme_update_ns_info nvme: clean up the check for too large logic block sizes nvme: freeze the queue over ->lba_shift updates nvme: factor out a nvme_configure_metadata helper ...
Diffstat (limited to 'drivers/block')
-rw-r--r--drivers/block/null_blk.h5
-rw-r--r--drivers/block/null_blk_main.c22
-rw-r--r--drivers/block/null_blk_zoned.c319
-rw-r--r--drivers/block/rsxx/core.c2
4 files changed, 289 insertions, 59 deletions
diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h
index daed4a9c3436..d2e7db43a52a 100644
--- a/drivers/block/null_blk.h
+++ b/drivers/block/null_blk.h
@@ -42,6 +42,9 @@ struct nullb_device {
struct badblocks badblocks;
unsigned int nr_zones;
+ unsigned int nr_zones_imp_open;
+ unsigned int nr_zones_exp_open;
+ unsigned int nr_zones_closed;
struct blk_zone *zones;
sector_t zone_size_sects;
@@ -51,6 +54,8 @@ struct nullb_device {
unsigned long zone_size; /* zone size in MB if device is zoned */
unsigned long zone_capacity; /* zone capacity in MB if device is zoned */
unsigned int zone_nr_conv; /* number of conventional zones */
+ unsigned int zone_max_open; /* max number of open zones */
+ unsigned int zone_max_active; /* max number of active zones */
unsigned int submit_queues; /* number of submission queues */
unsigned int home_node; /* home node for the device */
unsigned int queue_mode; /* block interface */
diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c
index d74443a9c8fa..4685ea401d5b 100644
--- a/drivers/block/null_blk_main.c
+++ b/drivers/block/null_blk_main.c
@@ -164,6 +164,10 @@ static bool shared_tags;
module_param(shared_tags, bool, 0444);
MODULE_PARM_DESC(shared_tags, "Share tag set between devices for blk-mq");
+static bool g_shared_tag_bitmap;
+module_param_named(shared_tag_bitmap, g_shared_tag_bitmap, bool, 0444);
+MODULE_PARM_DESC(shared_tag_bitmap, "Use shared tag bitmap for all submission queues for blk-mq");
+
static int g_irqmode = NULL_IRQ_SOFTIRQ;
static int null_set_irqmode(const char *str, const struct kernel_param *kp)
@@ -208,6 +212,14 @@ static unsigned int g_zone_nr_conv;
module_param_named(zone_nr_conv, g_zone_nr_conv, uint, 0444);
MODULE_PARM_DESC(zone_nr_conv, "Number of conventional zones when block device is zoned. Default: 0");
+static unsigned int g_zone_max_open;
+module_param_named(zone_max_open, g_zone_max_open, uint, 0444);
+MODULE_PARM_DESC(zone_max_open, "Maximum number of open zones when block device is zoned. Default: 0 (no limit)");
+
+static unsigned int g_zone_max_active;
+module_param_named(zone_max_active, g_zone_max_active, uint, 0444);
+MODULE_PARM_DESC(zone_max_active, "Maximum number of active zones when block device is zoned. Default: 0 (no limit)");
+
static struct nullb_device *null_alloc_dev(void);
static void null_free_dev(struct nullb_device *dev);
static void null_del_dev(struct nullb *nullb);
@@ -347,6 +359,8 @@ NULLB_DEVICE_ATTR(zoned, bool, NULL);
NULLB_DEVICE_ATTR(zone_size, ulong, NULL);
NULLB_DEVICE_ATTR(zone_capacity, ulong, NULL);
NULLB_DEVICE_ATTR(zone_nr_conv, uint, NULL);
+NULLB_DEVICE_ATTR(zone_max_open, uint, NULL);
+NULLB_DEVICE_ATTR(zone_max_active, uint, NULL);
static ssize_t nullb_device_power_show(struct config_item *item, char *page)
{
@@ -464,6 +478,8 @@ static struct configfs_attribute *nullb_device_attrs[] = {
&nullb_device_attr_zone_size,
&nullb_device_attr_zone_capacity,
&nullb_device_attr_zone_nr_conv,
+ &nullb_device_attr_zone_max_open,
+ &nullb_device_attr_zone_max_active,
NULL,
};
@@ -517,7 +533,7 @@ nullb_group_drop_item(struct config_group *group, struct config_item *item)
static ssize_t memb_group_features_show(struct config_item *item, char *page)
{
return snprintf(page, PAGE_SIZE,
- "memory_backed,discard,bandwidth,cache,badblocks,zoned,zone_size,zone_capacity,zone_nr_conv\n");
+ "memory_backed,discard,bandwidth,cache,badblocks,zoned,zone_size,zone_capacity,zone_nr_conv,zone_max_open,zone_max_active\n");
}
CONFIGFS_ATTR_RO(memb_group_, features);
@@ -580,6 +596,8 @@ static struct nullb_device *null_alloc_dev(void)
dev->zone_size = g_zone_size;
dev->zone_capacity = g_zone_capacity;
dev->zone_nr_conv = g_zone_nr_conv;
+ dev->zone_max_open = g_zone_max_open;
+ dev->zone_max_active = g_zone_max_active;
return dev;
}
@@ -1692,6 +1710,8 @@ static int null_init_tag_set(struct nullb *nullb, struct blk_mq_tag_set *set)
set->flags = BLK_MQ_F_SHOULD_MERGE;
if (g_no_sched)
set->flags |= BLK_MQ_F_NO_SCHED;
+ if (g_shared_tag_bitmap)
+ set->flags |= BLK_MQ_F_TAG_HCTX_SHARED;
set->driver_data = NULL;
if ((nullb && nullb->dev->blocking) || g_blocking)
diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c
index 3d25c9ad2383..fa0cc70f05e6 100644
--- a/drivers/block/null_blk_zoned.c
+++ b/drivers/block/null_blk_zoned.c
@@ -51,6 +51,22 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
dev->zone_nr_conv);
}
+ /* Max active zones has to be < nbr of seq zones in order to be enforceable */
+ if (dev->zone_max_active >= dev->nr_zones - dev->zone_nr_conv) {
+ dev->zone_max_active = 0;
+ pr_info("zone_max_active limit disabled, limit >= zone count\n");
+ }
+
+ /* Max open zones has to be <= max active zones */
+ if (dev->zone_max_active && dev->zone_max_open > dev->zone_max_active) {
+ dev->zone_max_open = dev->zone_max_active;
+ pr_info("changed the maximum number of open zones to %u\n",
+ dev->nr_zones);
+ } else if (dev->zone_max_open >= dev->nr_zones - dev->zone_nr_conv) {
+ dev->zone_max_open = 0;
+ pr_info("zone_max_open limit disabled, limit >= zone count\n");
+ }
+
for (i = 0; i < dev->zone_nr_conv; i++) {
struct blk_zone *zone = &dev->zones[i];
@@ -99,6 +115,8 @@ int null_register_zoned_dev(struct nullb *nullb)
}
blk_queue_max_zone_append_sectors(q, dev->zone_size_sects);
+ blk_queue_max_open_zones(q, dev->zone_max_open);
+ blk_queue_max_active_zones(q, dev->zone_max_active);
return 0;
}
@@ -159,6 +177,103 @@ size_t null_zone_valid_read_len(struct nullb *nullb,
return (zone->wp - sector) << SECTOR_SHIFT;
}
+static blk_status_t null_close_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+ if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+ return BLK_STS_IOERR;
+
+ switch (zone->cond) {
+ case BLK_ZONE_COND_CLOSED:
+ /* close operation on closed is not an error */
+ return BLK_STS_OK;
+ case BLK_ZONE_COND_IMP_OPEN:
+ dev->nr_zones_imp_open--;
+ break;
+ case BLK_ZONE_COND_EXP_OPEN:
+ dev->nr_zones_exp_open--;
+ break;
+ case BLK_ZONE_COND_EMPTY:
+ case BLK_ZONE_COND_FULL:
+ default:
+ return BLK_STS_IOERR;
+ }
+
+ if (zone->wp == zone->start) {
+ zone->cond = BLK_ZONE_COND_EMPTY;
+ } else {
+ zone->cond = BLK_ZONE_COND_CLOSED;
+ dev->nr_zones_closed++;
+ }
+
+ return BLK_STS_OK;
+}
+
+static void null_close_first_imp_zone(struct nullb_device *dev)
+{
+ unsigned int i;
+
+ for (i = dev->zone_nr_conv; i < dev->nr_zones; i++) {
+ if (dev->zones[i].cond == BLK_ZONE_COND_IMP_OPEN) {
+ null_close_zone(dev, &dev->zones[i]);
+ return;
+ }
+ }
+}
+
+static bool null_can_set_active(struct nullb_device *dev)
+{
+ if (!dev->zone_max_active)
+ return true;
+
+ return dev->nr_zones_exp_open + dev->nr_zones_imp_open +
+ dev->nr_zones_closed < dev->zone_max_active;
+}
+
+static bool null_can_open(struct nullb_device *dev)
+{
+ if (!dev->zone_max_open)
+ return true;
+
+ if (dev->nr_zones_exp_open + dev->nr_zones_imp_open < dev->zone_max_open)
+ return true;
+
+ if (dev->nr_zones_imp_open && null_can_set_active(dev)) {
+ null_close_first_imp_zone(dev);
+ return true;
+ }
+
+ return false;
+}
+
+/*
+ * This function matches the manage open zone resources function in the ZBC standard,
+ * with the addition of max active zones support (added in the ZNS standard).
+ *
+ * The function determines if a zone can transition to implicit open or explicit open,
+ * while maintaining the max open zone (and max active zone) limit(s). It may close an
+ * implicit open zone in order to make additional zone resources available.
+ *
+ * ZBC states that an implicit open zone shall be closed only if there is not
+ * room within the open limit. However, with the addition of an active limit,
+ * it is not certain that closing an implicit open zone will allow a new zone
+ * to be opened, since we might already be at the active limit capacity.
+ */
+static bool null_has_zone_resources(struct nullb_device *dev, struct blk_zone *zone)
+{
+ switch (zone->cond) {
+ case BLK_ZONE_COND_EMPTY:
+ if (!null_can_set_active(dev))
+ return false;
+ fallthrough;
+ case BLK_ZONE_COND_CLOSED:
+ return null_can_open(dev);
+ default:
+ /* Should never be called for other states */
+ WARN_ON(1);
+ return false;
+ }
+}
+
static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
unsigned int nr_sectors, bool append)
{
@@ -177,43 +292,155 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
/* Cannot write to a full zone */
return BLK_STS_IOERR;
case BLK_ZONE_COND_EMPTY:
+ case BLK_ZONE_COND_CLOSED:
+ if (!null_has_zone_resources(dev, zone))
+ return BLK_STS_IOERR;
+ break;
case BLK_ZONE_COND_IMP_OPEN:
case BLK_ZONE_COND_EXP_OPEN:
+ break;
+ default:
+ /* Invalid zone condition */
+ return BLK_STS_IOERR;
+ }
+
+ /*
+ * Regular writes must be at the write pointer position.
+ * Zone append writes are automatically issued at the write
+ * pointer and the position returned using the request or BIO
+ * sector.
+ */
+ if (append) {
+ sector = zone->wp;
+ if (cmd->bio)
+ cmd->bio->bi_iter.bi_sector = sector;
+ else
+ cmd->rq->__sector = sector;
+ } else if (sector != zone->wp) {
+ return BLK_STS_IOERR;
+ }
+
+ if (zone->wp + nr_sectors > zone->start + zone->capacity)
+ return BLK_STS_IOERR;
+
+ if (zone->cond == BLK_ZONE_COND_CLOSED) {
+ dev->nr_zones_closed--;
+ dev->nr_zones_imp_open++;
+ } else if (zone->cond == BLK_ZONE_COND_EMPTY) {
+ dev->nr_zones_imp_open++;
+ }
+ if (zone->cond != BLK_ZONE_COND_EXP_OPEN)
+ zone->cond = BLK_ZONE_COND_IMP_OPEN;
+
+ ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
+ if (ret != BLK_STS_OK)
+ return ret;
+
+ zone->wp += nr_sectors;
+ if (zone->wp == zone->start + zone->capacity) {
+ if (zone->cond == BLK_ZONE_COND_EXP_OPEN)
+ dev->nr_zones_exp_open--;
+ else if (zone->cond == BLK_ZONE_COND_IMP_OPEN)
+ dev->nr_zones_imp_open--;
+ zone->cond = BLK_ZONE_COND_FULL;
+ }
+ return BLK_STS_OK;
+}
+
+static blk_status_t null_open_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+ if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+ return BLK_STS_IOERR;
+
+ switch (zone->cond) {
+ case BLK_ZONE_COND_EXP_OPEN:
+ /* open operation on exp open is not an error */
+ return BLK_STS_OK;
+ case BLK_ZONE_COND_EMPTY:
+ if (!null_has_zone_resources(dev, zone))
+ return BLK_STS_IOERR;
+ break;
+ case BLK_ZONE_COND_IMP_OPEN:
+ dev->nr_zones_imp_open--;
+ break;
case BLK_ZONE_COND_CLOSED:
- /*
- * Regular writes must be at the write pointer position.
- * Zone append writes are automatically issued at the write
- * pointer and the position returned using the request or BIO
- * sector.
- */
- if (append) {
- sector = zone->wp;
- if (cmd->bio)
- cmd->bio->bi_iter.bi_sector = sector;
- else
- cmd->rq->__sector = sector;
- } else if (sector != zone->wp) {
+ if (!null_has_zone_resources(dev, zone))
return BLK_STS_IOERR;
- }
+ dev->nr_zones_closed--;
+ break;
+ case BLK_ZONE_COND_FULL:
+ default:
+ return BLK_STS_IOERR;
+ }
+
+ zone->cond = BLK_ZONE_COND_EXP_OPEN;
+ dev->nr_zones_exp_open++;
- if (zone->wp + nr_sectors > zone->start + zone->capacity)
+ return BLK_STS_OK;
+}
+
+static blk_status_t null_finish_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+ if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+ return BLK_STS_IOERR;
+
+ switch (zone->cond) {
+ case BLK_ZONE_COND_FULL:
+ /* finish operation on full is not an error */
+ return BLK_STS_OK;
+ case BLK_ZONE_COND_EMPTY:
+ if (!null_has_zone_resources(dev, zone))
return BLK_STS_IOERR;
+ break;
+ case BLK_ZONE_COND_IMP_OPEN:
+ dev->nr_zones_imp_open--;
+ break;
+ case BLK_ZONE_COND_EXP_OPEN:
+ dev->nr_zones_exp_open--;
+ break;
+ case BLK_ZONE_COND_CLOSED:
+ if (!null_has_zone_resources(dev, zone))
+ return BLK_STS_IOERR;
+ dev->nr_zones_closed--;
+ break;
+ default:
+ return BLK_STS_IOERR;
+ }
- if (zone->cond != BLK_ZONE_COND_EXP_OPEN)
- zone->cond = BLK_ZONE_COND_IMP_OPEN;
+ zone->cond = BLK_ZONE_COND_FULL;
+ zone->wp = zone->start + zone->len;
- ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
- if (ret != BLK_STS_OK)
- return ret;
+ return BLK_STS_OK;
+}
+
+static blk_status_t null_reset_zone(struct nullb_device *dev, struct blk_zone *zone)
+{
+ if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+ return BLK_STS_IOERR;
- zone->wp += nr_sectors;
- if (zone->wp == zone->start + zone->capacity)
- zone->cond = BLK_ZONE_COND_FULL;
+ switch (zone->cond) {
+ case BLK_ZONE_COND_EMPTY:
+ /* reset operation on empty is not an error */
return BLK_STS_OK;
+ case BLK_ZONE_COND_IMP_OPEN:
+ dev->nr_zones_imp_open--;
+ break;
+ case BLK_ZONE_COND_EXP_OPEN:
+ dev->nr_zones_exp_open--;
+ break;
+ case BLK_ZONE_COND_CLOSED:
+ dev->nr_zones_closed--;
+ break;
+ case BLK_ZONE_COND_FULL:
+ break;
default:
- /* Invalid zone condition */
return BLK_STS_IOERR;
}
+
+ zone->cond = BLK_ZONE_COND_EMPTY;
+ zone->wp = zone->start;
+
+ return BLK_STS_OK;
}
static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
@@ -222,56 +449,34 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
struct nullb_device *dev = cmd->nq->dev;
unsigned int zone_no = null_zone_no(dev, sector);
struct blk_zone *zone = &dev->zones[zone_no];
+ blk_status_t ret = BLK_STS_OK;
size_t i;
switch (op) {
case REQ_OP_ZONE_RESET_ALL:
- for (i = 0; i < dev->nr_zones; i++) {
- if (zone[i].type == BLK_ZONE_TYPE_CONVENTIONAL)
- continue;
- zone[i].cond = BLK_ZONE_COND_EMPTY;
- zone[i].wp = zone[i].start;
- }
+ for (i = dev->zone_nr_conv; i < dev->nr_zones; i++)
+ null_reset_zone(dev, &dev->zones[i]);
break;
case REQ_OP_ZONE_RESET:
- if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
- return BLK_STS_IOERR;
-
- zone->cond = BLK_ZONE_COND_EMPTY;
- zone->wp = zone->start;
+ ret = null_reset_zone(dev, zone);
break;
case REQ_OP_ZONE_OPEN:
- if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
- return BLK_STS_IOERR;
- if (zone->cond == BLK_ZONE_COND_FULL)
- return BLK_STS_IOERR;
-
- zone->cond = BLK_ZONE_COND_EXP_OPEN;
+ ret = null_open_zone(dev, zone);
break;
case REQ_OP_ZONE_CLOSE:
- if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
- return BLK_STS_IOERR;
- if (zone->cond == BLK_ZONE_COND_FULL)
- return BLK_STS_IOERR;
-
- if (zone->wp == zone->start)
- zone->cond = BLK_ZONE_COND_EMPTY;
- else
- zone->cond = BLK_ZONE_COND_CLOSED;
+ ret = null_close_zone(dev, zone);
break;
case REQ_OP_ZONE_FINISH:
- if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
- return BLK_STS_IOERR;
-
- zone->cond = BLK_ZONE_COND_FULL;
- zone->wp = zone->start + zone->len;
+ ret = null_finish_zone(dev, zone);
break;
default:
return BLK_STS_NOTSUPP;
}
- trace_nullb_zone_op(cmd, zone_no, zone->cond);
- return BLK_STS_OK;
+ if (ret == BLK_STS_OK)
+ trace_nullb_zone_op(cmd, zone_no, zone->cond);
+
+ return ret;
}
blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op,
diff --git a/drivers/block/rsxx/core.c b/drivers/block/rsxx/core.c
index 8799e3bab067..63f549889f87 100644
--- a/drivers/block/rsxx/core.c
+++ b/drivers/block/rsxx/core.c
@@ -439,7 +439,7 @@ static void card_state_change(struct rsxx_cardinfo *card,
case CARD_STATE_FAULT:
dev_crit(CARD_TO_DEV(card),
"Hardware Fault reported!\n");
- /* Fall through. */
+ fallthrough;
/* Everything else, detach DMA interface if it's attached. */
case CARD_STATE_SHUTDOWN: