summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJens Axboe <axboe@kernel.dk>2020-05-09 16:13:58 -0600
committerJens Axboe <axboe@kernel.dk>2020-05-09 16:13:58 -0600
commit873f1c8df7e0dac2156a722f3c7848a30693a242 (patch)
tree629e95d72af5be4cfd843e2c369b59c3de5baae6
parent8b075e5ba459c7afdd7b2fde14cbc01c51e25eac (diff)
parent59c7c3caaaf8750df4ec3255082f15eb4e371514 (diff)
Merge branch 'block-5.7' into for-5.8/block
Pull in block-5.7 fixes for 5.8. Mostly to resolve a conflict with the blk-iocost changes, but we also need the base of the bdi use-after-free as well as we build on top of it. * block-5.7: nvme: fix possible hang when ns scanning fails during error recovery nvme-pci: fix "slimmer CQ head update" bdi: add a ->dev_name field to struct backing_dev_info bdi: use bdi_dev_name() to get device name bdi: move bdi_dev_name out of line vboxsf: don't use the source name in the bdi name iocost: protect iocg->abs_vdebt with iocg->waitq.lock block: remove the bd_openers checks in blk_drop_partitions nvme: prevent double free in nvme_alloc_ns() error handling null_blk: Cleanup zoned device initialization null_blk: Fix zoned command handling block: remove unused header blk-iocost: Fix error on iocost_ioc_vrate_adj bdev: Reduce time holding bd_mutex in sync in blkdev_close() buffer: remove useless comment and WB_REASON_FREE_MORE_MEM, reason. Signed-off-by: Jens Axboe <axboe@kernel.dk>
-rw-r--r--block/bfq-iosched.c6
-rw-r--r--block/blk-cgroup.c2
-rw-r--r--block/blk-iocost.c121
-rw-r--r--block/partitions/core.c2
-rw-r--r--drivers/block/null_blk.h29
-rw-r--r--drivers/block/null_blk_main.c62
-rw-r--r--drivers/block/null_blk_zoned.c45
-rw-r--r--drivers/nvme/host/core.c4
-rw-r--r--drivers/nvme/host/pci.c6
-rw-r--r--fs/block_dev.c11
-rw-r--r--fs/buffer.c2
-rw-r--r--fs/ceph/debugfs.c2
-rw-r--r--fs/vboxsf/super.c2
-rw-r--r--include/linux/backing-dev-defs.h2
-rw-r--r--include/linux/backing-dev.h9
-rw-r--r--include/trace/events/iocost.h6
-rw-r--r--include/trace/events/wbt.h8
-rw-r--r--include/trace/events/writeback.h1
-rw-r--r--mm/backing-dev.c13
-rw-r--r--tools/cgroup/iocost_monitor.py7
20 files changed, 209 insertions, 131 deletions
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 78ba57efd16b..3d411716d7ee 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -123,6 +123,7 @@
#include <linux/ioprio.h>
#include <linux/sbitmap.h>
#include <linux/delay.h>
+#include <linux/backing-dev.h>
#include "blk.h"
#include "blk-mq.h"
@@ -4976,8 +4977,9 @@ bfq_set_next_ioprio_data(struct bfq_queue *bfqq, struct bfq_io_cq *bic)
ioprio_class = IOPRIO_PRIO_CLASS(bic->ioprio);
switch (ioprio_class) {
default:
- dev_err(bfqq->bfqd->queue->backing_dev_info->dev,
- "bfq: bad prio class %d\n", ioprio_class);
+ pr_err("bdi %s: bfq: bad prio class %d\n",
+ bdi_dev_name(bfqq->bfqd->queue->backing_dev_info),
+ ioprio_class);
/* fall through */
case IOPRIO_CLASS_NONE:
/*
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 0a63c6cbbcb1..0ecc897b225c 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -496,7 +496,7 @@ const char *blkg_dev_name(struct blkcg_gq *blkg)
{
/* some drivers (floppy) instantiate a queue w/o disk registered */
if (blkg->q->backing_dev_info->dev)
- return dev_name(blkg->q->backing_dev_info->dev);
+ return bdi_dev_name(blkg->q->backing_dev_info);
return NULL;
}
diff --git a/block/blk-iocost.c b/block/blk-iocost.c
index 9a667dd75eef..12eefb3d113f 100644
--- a/block/blk-iocost.c
+++ b/block/blk-iocost.c
@@ -467,7 +467,7 @@ struct ioc_gq {
*/
atomic64_t vtime;
atomic64_t done_vtime;
- atomic64_t abs_vdebt;
+ u64 abs_vdebt;
u64 last_vtime;
/*
@@ -1143,7 +1143,7 @@ static void iocg_kick_waitq(struct ioc_gq *iocg, struct ioc_now *now)
struct iocg_wake_ctx ctx = { .iocg = iocg };
u64 margin_ns = (u64)(ioc->period_us *
WAITQ_TIMER_MARGIN_PCT / 100) * NSEC_PER_USEC;
- u64 abs_vdebt, vdebt, vshortage, expires, oexpires;
+ u64 vdebt, vshortage, expires, oexpires;
s64 vbudget;
u32 hw_inuse;
@@ -1153,18 +1153,15 @@ static void iocg_kick_waitq(struct ioc_gq *iocg, struct ioc_now *now)
vbudget = now->vnow - atomic64_read(&iocg->vtime);
/* pay off debt */
- abs_vdebt = atomic64_read(&iocg->abs_vdebt);
- vdebt = abs_cost_to_cost(abs_vdebt, hw_inuse);
+ vdebt = abs_cost_to_cost(iocg->abs_vdebt, hw_inuse);
if (vdebt && vbudget > 0) {
u64 delta = min_t(u64, vbudget, vdebt);
u64 abs_delta = min(cost_to_abs_cost(delta, hw_inuse),
- abs_vdebt);
+ iocg->abs_vdebt);
atomic64_add(delta, &iocg->vtime);
atomic64_add(delta, &iocg->done_vtime);
- atomic64_sub(abs_delta, &iocg->abs_vdebt);
- if (WARN_ON_ONCE(atomic64_read(&iocg->abs_vdebt) < 0))
- atomic64_set(&iocg->abs_vdebt, 0);
+ iocg->abs_vdebt -= abs_delta;
}
/*
@@ -1220,12 +1217,18 @@ static bool iocg_kick_delay(struct ioc_gq *iocg, struct ioc_now *now)
u64 delta_ns, expires, oexpires;
u32 hw_inuse;
+ lockdep_assert_held(&iocg->waitq.lock);
+
/* debt-adjust vtime */
current_hweight(iocg, NULL, &hw_inuse);
- vtime += abs_cost_to_cost(atomic64_read(&iocg->abs_vdebt), hw_inuse);
+ vtime += abs_cost_to_cost(iocg->abs_vdebt, hw_inuse);
- /* clear or maintain depending on the overage */
- if (time_before_eq64(vtime, now->vnow)) {
+ /*
+ * Clear or maintain depending on the overage. Non-zero vdebt is what
+ * guarantees that @iocg is online and future iocg_kick_delay() will
+ * clear use_delay. Don't leave it on when there's no vdebt.
+ */
+ if (!iocg->abs_vdebt || time_before_eq64(vtime, now->vnow)) {
blkcg_clear_delay(blkg);
return false;
}
@@ -1254,9 +1257,12 @@ static enum hrtimer_restart iocg_delay_timer_fn(struct hrtimer *timer)
{
struct ioc_gq *iocg = container_of(timer, struct ioc_gq, delay_timer);
struct ioc_now now;
+ unsigned long flags;
+ spin_lock_irqsave(&iocg->waitq.lock, flags);
ioc_now(iocg->ioc, &now);
iocg_kick_delay(iocg, &now);
+ spin_unlock_irqrestore(&iocg->waitq.lock, flags);
return HRTIMER_NORESTART;
}
@@ -1364,14 +1370,13 @@ static void ioc_timer_fn(struct timer_list *timer)
* should have woken up in the last period and expire idle iocgs.
*/
list_for_each_entry_safe(iocg, tiocg, &ioc->active_iocgs, active_list) {
- if (!waitqueue_active(&iocg->waitq) &&
- !atomic64_read(&iocg->abs_vdebt) && !iocg_is_idle(iocg))
+ if (!waitqueue_active(&iocg->waitq) && iocg->abs_vdebt &&
+ !iocg_is_idle(iocg))
continue;
spin_lock(&iocg->waitq.lock);
- if (waitqueue_active(&iocg->waitq) ||
- atomic64_read(&iocg->abs_vdebt)) {
+ if (waitqueue_active(&iocg->waitq) || iocg->abs_vdebt) {
/* might be oversleeping vtime / hweight changes, kick */
iocg_kick_waitq(iocg, &now);
iocg_kick_delay(iocg, &now);
@@ -1587,7 +1592,7 @@ skip_surplus_transfers:
vrate_min, vrate_max);
}
- trace_iocost_ioc_vrate_adj(ioc, vrate, &missed_ppm, rq_wait_pct,
+ trace_iocost_ioc_vrate_adj(ioc, vrate, missed_ppm, rq_wait_pct,
nr_lagging, nr_shortages,
nr_surpluses);
@@ -1596,7 +1601,7 @@ skip_surplus_transfers:
ioc->period_us * vrate * INUSE_MARGIN_PCT, 100);
} else if (ioc->busy_level != prev_busy_level || nr_lagging) {
trace_iocost_ioc_vrate_adj(ioc, atomic64_read(&ioc->vtime_rate),
- &missed_ppm, rq_wait_pct, nr_lagging,
+ missed_ppm, rq_wait_pct, nr_lagging,
nr_shortages, nr_surpluses);
}
@@ -1739,28 +1744,49 @@ static void ioc_rqos_throttle(struct rq_qos *rqos, struct bio *bio)
* tests are racy but the races aren't systemic - we only miss once
* in a while which is fine.
*/
- if (!waitqueue_active(&iocg->waitq) &&
- !atomic64_read(&iocg->abs_vdebt) &&
+ if (!waitqueue_active(&iocg->waitq) && !iocg->abs_vdebt &&
time_before_eq64(vtime + cost, now.vnow)) {
iocg_commit_bio(iocg, bio, cost);
return;
}
/*
- * We're over budget. If @bio has to be issued regardless,
- * remember the abs_cost instead of advancing vtime.
- * iocg_kick_waitq() will pay off the debt before waking more IOs.
+ * We activated above but w/o any synchronization. Deactivation is
+ * synchronized with waitq.lock and we won't get deactivated as long
+ * as we're waiting or has debt, so we're good if we're activated
+ * here. In the unlikely case that we aren't, just issue the IO.
+ */
+ spin_lock_irq(&iocg->waitq.lock);
+
+ if (unlikely(list_empty(&iocg->active_list))) {
+ spin_unlock_irq(&iocg->waitq.lock);
+ iocg_commit_bio(iocg, bio, cost);
+ return;
+ }
+
+ /*
+ * We're over budget. If @bio has to be issued regardless, remember
+ * the abs_cost instead of advancing vtime. iocg_kick_waitq() will pay
+ * off the debt before waking more IOs.
+ *
* This way, the debt is continuously paid off each period with the
- * actual budget available to the cgroup. If we just wound vtime,
- * we would incorrectly use the current hw_inuse for the entire
- * amount which, for example, can lead to the cgroup staying
- * blocked for a long time even with substantially raised hw_inuse.
+ * actual budget available to the cgroup. If we just wound vtime, we
+ * would incorrectly use the current hw_inuse for the entire amount
+ * which, for example, can lead to the cgroup staying blocked for a
+ * long time even with substantially raised hw_inuse.
+ *
+ * An iocg with vdebt should stay online so that the timer can keep
+ * deducting its vdebt and [de]activate use_delay mechanism
+ * accordingly. We don't want to race against the timer trying to
+ * clear them and leave @iocg inactive w/ dangling use_delay heavily
+ * penalizing the cgroup and its descendants.
*/
if (bio_issue_as_root_blkg(bio) || fatal_signal_pending(current)) {
- atomic64_add(abs_cost, &iocg->abs_vdebt);
+ iocg->abs_vdebt += abs_cost;
if (iocg_kick_delay(iocg, &now))
blkcg_schedule_throttle(rqos->q,
(bio->bi_opf & REQ_SWAP) == REQ_SWAP);
+ spin_unlock_irq(&iocg->waitq.lock);
return;
}
@@ -1777,20 +1803,6 @@ static void ioc_rqos_throttle(struct rq_qos *rqos, struct bio *bio)
* All waiters are on iocg->waitq and the wait states are
* synchronized using waitq.lock.
*/
- spin_lock_irq(&iocg->waitq.lock);
-
- /*
- * We activated above but w/o any synchronization. Deactivation is
- * synchronized with waitq.lock and we won't get deactivated as
- * long as we're waiting, so we're good if we're activated here.
- * In the unlikely case that we are deactivated, just issue the IO.
- */
- if (unlikely(list_empty(&iocg->active_list))) {
- spin_unlock_irq(&iocg->waitq.lock);
- iocg_commit_bio(iocg, bio, cost);
- return;
- }
-
init_waitqueue_func_entry(&wait.wait, iocg_wake_fn);
wait.wait.private = current;
wait.bio = bio;
@@ -1822,6 +1834,7 @@ static void ioc_rqos_merge(struct rq_qos *rqos, struct request *rq,
struct ioc_now now;
u32 hw_inuse;
u64 abs_cost, cost;
+ unsigned long flags;
/* bypass if disabled or for root cgroup */
if (!ioc->enabled || !iocg->level)
@@ -1841,15 +1854,28 @@ static void ioc_rqos_merge(struct rq_qos *rqos, struct request *rq,
iocg->cursor = bio_end;
/*
- * Charge if there's enough vtime budget and the existing request
- * has cost assigned. Otherwise, account it as debt. See debt
- * handling in ioc_rqos_throttle() for details.
+ * Charge if there's enough vtime budget and the existing request has
+ * cost assigned.
*/
if (rq->bio && rq->bio->bi_iocost_cost &&
- time_before_eq64(atomic64_read(&iocg->vtime) + cost, now.vnow))
+ time_before_eq64(atomic64_read(&iocg->vtime) + cost, now.vnow)) {
iocg_commit_bio(iocg, bio, cost);
- else
- atomic64_add(abs_cost, &iocg->abs_vdebt);
+ return;
+ }
+
+ /*
+ * Otherwise, account it as debt if @iocg is online, which it should
+ * be for the vast majority of cases. See debt handling in
+ * ioc_rqos_throttle() for details.
+ */
+ spin_lock_irqsave(&iocg->waitq.lock, flags);
+ if (likely(!list_empty(&iocg->active_list))) {
+ iocg->abs_vdebt += abs_cost;
+ iocg_kick_delay(iocg, &now);
+ } else {
+ iocg_commit_bio(iocg, bio, cost);
+ }
+ spin_unlock_irqrestore(&iocg->waitq.lock, flags);
}
static void ioc_rqos_done_bio(struct rq_qos *rqos, struct bio *bio)
@@ -2021,7 +2047,6 @@ static void ioc_pd_init(struct blkg_policy_data *pd)
iocg->ioc = ioc;
atomic64_set(&iocg->vtime, now.vnow);
atomic64_set(&iocg->done_vtime, now.vnow);
- atomic64_set(&iocg->abs_vdebt, 0);
atomic64_set(&iocg->active_period, atomic64_read(&ioc->cur_period));
INIT_LIST_HEAD(&iocg->active_list);
iocg->hweight_active = HWEIGHT_WHOLE;
diff --git a/block/partitions/core.c b/block/partitions/core.c
index c085bf85509b..873999e2e2f2 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -610,7 +610,7 @@ int blk_drop_partitions(struct block_device *bdev)
if (!disk_part_scan_enabled(bdev->bd_disk))
return 0;
- if (bdev->bd_part_count || bdev->bd_openers > 1)
+ if (bdev->bd_part_count)
return -EBUSY;
sync_blockdev(bdev);
diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h
index 62b660821dbc..81b311c9d781 100644
--- a/drivers/block/null_blk.h
+++ b/drivers/block/null_blk.h
@@ -85,26 +85,35 @@ struct nullb {
char disk_name[DISK_NAME_LEN];
};
+blk_status_t null_process_cmd(struct nullb_cmd *cmd,
+ enum req_opf op, sector_t sector,
+ unsigned int nr_sectors);
+
#ifdef CONFIG_BLK_DEV_ZONED
-int null_zone_init(struct nullb_device *dev);
-void null_zone_exit(struct nullb_device *dev);
+int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q);
+int null_register_zoned_dev(struct nullb *nullb);
+void null_free_zoned_dev(struct nullb_device *dev);
int null_report_zones(struct gendisk *disk, sector_t sector,
unsigned int nr_zones, report_zones_cb cb, void *data);
-blk_status_t null_handle_zoned(struct nullb_cmd *cmd,
- enum req_opf op, sector_t sector,
- sector_t nr_sectors);
+blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd,
+ enum req_opf op, sector_t sector,
+ sector_t nr_sectors);
size_t null_zone_valid_read_len(struct nullb *nullb,
sector_t sector, unsigned int len);
#else
-static inline int null_zone_init(struct nullb_device *dev)
+static inline int null_init_zoned_dev(struct nullb_device *dev,
+ struct request_queue *q)
{
pr_err("CONFIG_BLK_DEV_ZONED not enabled\n");
return -EINVAL;
}
-static inline void null_zone_exit(struct nullb_device *dev) {}
-static inline blk_status_t null_handle_zoned(struct nullb_cmd *cmd,
- enum req_opf op, sector_t sector,
- sector_t nr_sectors)
+static inline int null_register_zoned_dev(struct nullb *nullb)
+{
+ return -ENODEV;
+}
+static inline void null_free_zoned_dev(struct nullb_device *dev) {}
+static inline blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd,
+ enum req_opf op, sector_t sector, sector_t nr_sectors)
{
return BLK_STS_NOTSUPP;
}
diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c
index 4e1c0712278e..8efd8778e209 100644
--- a/drivers/block/null_blk_main.c
+++ b/drivers/block/null_blk_main.c
@@ -580,7 +580,7 @@ static void null_free_dev(struct nullb_device *dev)
if (!dev)
return;
- null_zone_exit(dev);
+ null_free_zoned_dev(dev);
badblocks_exit(&dev->badblocks);
kfree(dev);
}
@@ -1276,6 +1276,25 @@ static inline void nullb_complete_cmd(struct nullb_cmd *cmd)
}
}
+blk_status_t null_process_cmd(struct nullb_cmd *cmd,
+ enum req_opf op, sector_t sector,
+ unsigned int nr_sectors)
+{
+ struct nullb_device *dev = cmd->nq->dev;
+ blk_status_t ret;
+
+ if (dev->badblocks.shift != -1) {
+ ret = null_handle_badblocks(cmd, sector, nr_sectors);
+ if (ret != BLK_STS_OK)
+ return ret;
+ }
+
+ if (dev->memory_backed)
+ return null_handle_memory_backed(cmd, op);
+
+ return BLK_STS_OK;
+}
+
static blk_status_t null_handle_cmd(struct nullb_cmd *cmd, sector_t sector,
sector_t nr_sectors, enum req_opf op)
{
@@ -1294,17 +1313,11 @@ static blk_status_t null_handle_cmd(struct nullb_cmd *cmd, sector_t sector,
goto out;
}
- if (nullb->dev->badblocks.shift != -1) {
- cmd->error = null_handle_badblocks(cmd, sector, nr_sectors);
- if (cmd->error != BLK_STS_OK)
- goto out;
- }
-
- if (dev->memory_backed)
- cmd->error = null_handle_memory_backed(cmd, op);
-
- if (!cmd->error && dev->zoned)
- cmd->error = null_handle_zoned(cmd, op, sector, nr_sectors);
+ if (dev->zoned)
+ cmd->error = null_process_zoned_cmd(cmd, op,
+ sector, nr_sectors);
+ else
+ cmd->error = null_process_cmd(cmd, op, sector, nr_sectors);
out:
nullb_complete_cmd(cmd);
@@ -1605,19 +1618,12 @@ static int null_gendisk_register(struct nullb *nullb)
disk->queue = nullb->q;
strncpy(disk->disk_name, nullb->disk_name, DISK_NAME_LEN);
-#ifdef CONFIG_BLK_DEV_ZONED
if (nullb->dev->zoned) {
- if (queue_is_mq(nullb->q)) {
- int ret = blk_revalidate_disk_zones(disk);
- if (ret)
- return ret;
- } else {
- blk_queue_chunk_sectors(nullb->q,
- nullb->dev->zone_size_sects);
- nullb->q->nr_zones = blkdev_nr_zones(disk);
- }
+ int ret = null_register_zoned_dev(nullb);
+
+ if (ret)
+ return ret;
}
-#endif
add_disk(disk);
return 0;
@@ -1773,14 +1779,9 @@ static int null_add_dev(struct nullb_device *dev)
}
if (dev->zoned) {
- rv = null_zone_init(dev);
+ rv = null_init_zoned_dev(dev, nullb->q);
if (rv)
goto out_cleanup_blk_queue;
-
- nullb->q->limits.zoned = BLK_ZONED_HM;
- blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, nullb->q);
- blk_queue_required_elevator_features(nullb->q,
- ELEVATOR_F_ZBD_SEQ_WRITE);
}
nullb->q->queuedata = nullb;
@@ -1809,8 +1810,7 @@ static int null_add_dev(struct nullb_device *dev)
return 0;
out_cleanup_zone:
- if (dev->zoned)
- null_zone_exit(dev);
+ null_free_zoned_dev(dev);
out_cleanup_blk_queue:
blk_cleanup_queue(nullb->q);
out_cleanup_tags:
diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c
index 673618d8222a..9e4bcdad1a80 100644
--- a/drivers/block/null_blk_zoned.c
+++ b/drivers/block/null_blk_zoned.c
@@ -13,7 +13,7 @@ static inline unsigned int null_zone_no(struct nullb_device *dev, sector_t sect)
return sect >> ilog2(dev->zone_size_sects);
}
-int null_zone_init(struct nullb_device *dev)
+int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
{
sector_t dev_size = (sector_t)dev->size * 1024 * 1024;
sector_t sector = 0;
@@ -61,10 +61,27 @@ int null_zone_init(struct nullb_device *dev)
sector += dev->zone_size_sects;
}
+ q->limits.zoned = BLK_ZONED_HM;
+ blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
+ blk_queue_required_elevator_features(q, ELEVATOR_F_ZBD_SEQ_WRITE);
+
+ return 0;
+}
+
+int null_register_zoned_dev(struct nullb *nullb)
+{
+ struct request_queue *q = nullb->q;
+
+ if (queue_is_mq(q))
+ return blk_revalidate_disk_zones(nullb->disk);
+
+ blk_queue_chunk_sectors(q, nullb->dev->zone_size_sects);
+ q->nr_zones = blkdev_nr_zones(nullb->disk);
+
return 0;
}
-void null_zone_exit(struct nullb_device *dev)
+void null_free_zoned_dev(struct nullb_device *dev)
{
kvfree(dev->zones);
}
@@ -126,11 +143,16 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
struct nullb_device *dev = cmd->nq->dev;
unsigned int zno = null_zone_no(dev, sector);
struct blk_zone *zone = &dev->zones[zno];
+ blk_status_t ret;
+
+ trace_nullb_zone_op(cmd, zno, zone->cond);
+
+ if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL)
+ return null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors);
switch (zone->cond) {
case BLK_ZONE_COND_FULL:
/* Cannot write to a full zone */
- cmd->error = BLK_STS_IOERR;
return BLK_STS_IOERR;
case BLK_ZONE_COND_EMPTY:
case BLK_ZONE_COND_IMP_OPEN:
@@ -143,19 +165,18 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
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->len)
zone->cond = BLK_ZONE_COND_FULL;
- break;
- case BLK_ZONE_COND_NOT_WP:
- break;
+ return BLK_STS_OK;
default:
/* Invalid zone condition */
return BLK_STS_IOERR;
}
-
- trace_nullb_zone_op(cmd, zno, zone->cond);
- return BLK_STS_OK;
}
static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
@@ -216,8 +237,8 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op,
return BLK_STS_OK;
}
-blk_status_t null_handle_zoned(struct nullb_cmd *cmd, enum req_opf op,
- sector_t sector, sector_t nr_sectors)
+blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op,
+ sector_t sector, sector_t nr_sectors)
{
switch (op) {
case REQ_OP_WRITE:
@@ -229,6 +250,6 @@ blk_status_t null_handle_zoned(struct nullb_cmd *cmd, enum req_opf op,
case REQ_OP_ZONE_FINISH:
return null_zone_mgmt(cmd, op, sector);
default:
- return BLK_STS_OK;
+ return null_process_cmd(cmd, op, sector, nr_sectors);
}
}
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 91c1bd659947..f3c037f5a9ba 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -1110,7 +1110,7 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid,
* Don't treat an error as fatal, as we potentially already
* have a NGUID or EUI-64.
*/
- if (status > 0)
+ if (status > 0 && !(status & NVME_SC_DNR))
status = 0;
goto free_data;
}
@@ -3642,6 +3642,8 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
return;
out_put_disk:
+ /* prevent double queue cleanup */
+ ns->disk->queue = NULL;
put_disk(ns->disk);
out_unlink_ns:
mutex_lock(&ctrl->subsys->lock);
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 4e79e412b276..e13c370de830 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -973,9 +973,13 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq, u16 idx)
static inline void nvme_update_cq_head(struct nvme_queue *nvmeq)
{
- if (++nvmeq->cq_head == nvmeq->q_depth) {
+ u16 tmp = nvmeq->cq_head + 1;
+
+ if (tmp == nvmeq->q_depth) {
nvmeq->cq_head = 0;
nvmeq->cq_phase ^= 1;
+ } else {
+ nvmeq->cq_head = tmp;
}
}
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 5eb30a474f6d..7cbb7b79935e 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -19,7 +19,6 @@
#include <linux/module.h>
#include <linux/blkpg.h>
#include <linux/magic.h>
-#include <linux/dax.h>
#include <linux/buffer_head.h>
#include <linux/swap.h>
#include <linux/pagevec.h>
@@ -1876,6 +1875,16 @@ static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part)
struct gendisk *disk = bdev->bd_disk;
struct block_device *victim = NULL;
+ /*
+ * Sync early if it looks like we're the last one. If someone else
+ * opens the block device between now and the decrement of bd_openers
+ * then we did a sync that we didn't need to, but that's not the end
+ * of the world and we want to avoid long (could be several minute)
+ * syncs while holding the mutex.
+ */
+ if (bdev->bd_openers == 1)
+ sync_blockdev(bdev);
+
mutex_lock_nested(&bdev->bd_mutex, for_part);
if (for_part)
bdev->bd_part_count--;
diff --git a/fs/buffer.c b/fs/buffer.c
index 599a0bf7257b..a60f60396cfa 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -967,7 +967,7 @@ grow_dev_page(struct block_device *bdev, sector_t block,
struct page *page;
struct buffer_head *bh;
sector_t end_block;
- int ret = 0; /* Will call free_more_memory() */
+ int ret = 0;
gfp_t gfp_mask;
gfp_mask = mapping_gfp_constraint(inode->i_mapping, ~__GFP_FS) | gfp;
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 481ac97b4d25..dcaed75de9e6 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -271,7 +271,7 @@ void ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
&congestion_kb_fops);
snprintf(name, sizeof(name), "../../bdi/%s",
- dev_name(fsc->sb->s_bdi->dev));
+ bdi_dev_name(fsc->sb->s_bdi));
fsc->debugfs_bdi =
debugfs_create_symlink("bdi",
fsc->client->debugfs_dir,
diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c
index 675e26989376..8fe03b4a0d2b 100644
--- a/fs/vboxsf/super.c
+++ b/fs/vboxsf/super.c
@@ -164,7 +164,7 @@ static int vboxsf_fill_super(struct super_block *sb, struct fs_context *fc)
goto fail_free;
}
- err = super_setup_bdi_name(sb, "vboxsf-%s.%d", fc->source, sbi->bdi_id);
+ err = super_setup_bdi_name(sb, "vboxsf-%d", sbi->bdi_id);
if (err)
goto fail_free;
diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h
index 4fc87dee005a..7367150f962a 100644
--- a/include/linux/backing-dev-defs.h
+++ b/include/linux/backing-dev-defs.h
@@ -54,7 +54,6 @@ enum wb_reason {
WB_REASON_SYNC,
WB_REASON_PERIODIC,
WB_REASON_LAPTOP_TIMER,
- WB_REASON_FREE_MORE_MEM,
WB_REASON_FS_FREE_SPACE,
/*
* There is no bdi forker thread any more and works are done
@@ -220,6 +219,7 @@ struct backing_dev_info {
wait_queue_head_t wb_waitq;
struct device *dev;
+ char dev_name[64];
struct device *owner;
struct timer_list laptop_mode_wb_timer;
diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h
index f88197c1ffc2..c9ad5c3b7b4b 100644
--- a/include/linux/backing-dev.h
+++ b/include/linux/backing-dev.h
@@ -505,13 +505,6 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi)
(1 << WB_async_congested));
}
-extern const char *bdi_unknown_name;
-
-static inline const char *bdi_dev_name(struct backing_dev_info *bdi)
-{
- if (!bdi || !bdi->dev)
- return bdi_unknown_name;
- return dev_name(bdi->dev);
-}
+const char *bdi_dev_name(struct backing_dev_info *bdi);
#endif /* _LINUX_BACKING_DEV_H */
diff --git a/include/trace/events/iocost.h b/include/trace/events/iocost.h
index 7ecaa65b7106..c2f580fd371b 100644
--- a/include/trace/events/iocost.h
+++ b/include/trace/events/iocost.h
@@ -130,7 +130,7 @@ DEFINE_EVENT(iocg_inuse_update, iocost_inuse_reset,
TRACE_EVENT(iocost_ioc_vrate_adj,
- TP_PROTO(struct ioc *ioc, u64 new_vrate, u32 (*missed_ppm)[2],
+ TP_PROTO(struct ioc *ioc, u64 new_vrate, u32 *missed_ppm,
u32 rq_wait_pct, int nr_lagging, int nr_shortages,
int nr_surpluses),
@@ -155,8 +155,8 @@ TRACE_EVENT(iocost_ioc_vrate_adj,
__entry->old_vrate = atomic64_read(&ioc->vtime_rate);;
__entry->new_vrate = new_vrate;
__entry->busy_level = ioc->busy_level;
- __entry->read_missed_ppm = (*missed_ppm)[READ];
- __entry->write_missed_ppm = (*missed_ppm)[WRITE];
+ __entry->read_missed_ppm = missed_ppm[READ];
+ __entry->write_missed_ppm = missed_ppm[WRITE];
__entry->rq_wait_pct = rq_wait_pct;
__entry->nr_lagging = nr_lagging;
__entry->nr_shortages = nr_shortages;
diff --git a/include/trace/events/wbt.h b/include/trace/events/wbt.h
index 784814160197..9c66e59d859c 100644
--- a/include/trace/events/wbt.h
+++ b/include/trace/events/wbt.h
@@ -33,7 +33,7 @@ TRACE_EVENT(wbt_stat,
),
TP_fast_assign(
- strlcpy(__entry->name, dev_name(bdi->dev),
+ strlcpy(__entry->name, bdi_dev_name(bdi),
ARRAY_SIZE(__entry->name));
__entry->rmean = stat[0].mean;
__entry->rmin = stat[0].min;
@@ -68,7 +68,7 @@ TRACE_EVENT(wbt_lat,
),
TP_fast_assign(
- strlcpy(__entry->name, dev_name(bdi->dev),
+ strlcpy(__entry->name, bdi_dev_name(bdi),
ARRAY_SIZE(__entry->name));
__entry->lat = div_u64(lat, 1000);
),
@@ -105,7 +105,7 @@ TRACE_EVENT(wbt_step,
),
TP_fast_assign(
- strlcpy(__entry->name, dev_name(bdi->dev),
+ strlcpy(__entry->name, bdi_dev_name(bdi),
ARRAY_SIZE(__entry->name));
__entry->msg = msg;
__entry->step = step;
@@ -141,7 +141,7 @@ TRACE_EVENT(wbt_timer,
),
TP_fast_assign(
- strlcpy(__entry->name, dev_name(bdi->dev),
+ strlcpy(__entry->name, bdi_dev_name(bdi),
ARRAY_SIZE(__entry->name));
__entry->status = status;
__entry->step = step;
diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h
index d94def25e4dc..85a33bea76f1 100644
--- a/include/trace/events/writeback.h
+++ b/include/trace/events/writeback.h
@@ -36,7 +36,6 @@
EM( WB_REASON_SYNC, "sync") \
EM( WB_REASON_PERIODIC, "periodic") \
EM( WB_REASON_LAPTOP_TIMER, "laptop_timer") \
- EM( WB_REASON_FREE_MORE_MEM, "free_more_memory") \
EM( WB_REASON_FS_FREE_SPACE, "fs_free_space") \
EMe(WB_REASON_FORKER_THREAD, "forker_thread")
diff --git a/mm/backing-dev.c b/mm/backing-dev.c
index c81b4f3a7268..efc5b83acd2d 100644
--- a/mm/backing-dev.c
+++ b/mm/backing-dev.c
@@ -21,7 +21,7 @@ struct backing_dev_info noop_backing_dev_info = {
EXPORT_SYMBOL_GPL(noop_backing_dev_info);
static struct class *bdi_class;
-const char *bdi_unknown_name = "(unknown)";
+static const char *bdi_unknown_name = "(unknown)";
/*
* bdi_lock protects bdi_tree and updates to bdi_list. bdi_list has RCU
@@ -938,7 +938,8 @@ int bdi_register_va(struct backing_dev_info *bdi, const char *fmt, va_list args)
if (bdi->dev) /* The driver needs to use separate queues per device */
return 0;
- dev = device_create_vargs(bdi_class, NULL, MKDEV(0, 0), bdi, fmt, args);
+ vsnprintf(bdi->dev_name, sizeof(bdi->dev_name), fmt, args);
+ dev = device_create(bdi_class, NULL, MKDEV(0, 0), bdi, bdi->dev_name);
if (IS_ERR(dev))
return PTR_ERR(dev);
@@ -1043,6 +1044,14 @@ void bdi_put(struct backing_dev_info *bdi)
}
EXPORT_SYMBOL(bdi_put);
+const char *bdi_dev_name(struct backing_dev_info *bdi)
+{
+ if (!bdi || !bdi->dev)
+ return bdi_unknown_name;
+ return bdi->dev_name;
+}
+EXPORT_SYMBOL_GPL(bdi_dev_name);
+
static wait_queue_head_t congestion_wqh[2] = {
__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
diff --git a/tools/cgroup/iocost_monitor.py b/tools/cgroup/iocost_monitor.py
index 188b3379b9a1..3c21de88af9e 100644
--- a/tools/cgroup/iocost_monitor.py
+++ b/tools/cgroup/iocost_monitor.py
@@ -160,7 +160,12 @@ class IocgStat:
else:
self.inflight_pct = 0
- self.debt_ms = iocg.abs_vdebt.counter.value_() / VTIME_PER_USEC / 1000
+ # vdebt used to be an atomic64_t and is now u64, support both
+ try:
+ self.debt_ms = iocg.abs_vdebt.counter.value_() / VTIME_PER_USEC / 1000
+ except:
+ self.debt_ms = iocg.abs_vdebt.value_() / VTIME_PER_USEC / 1000
+
self.use_delay = blkg.use_delay.counter.value_()
self.delay_ms = blkg.delay_nsec.counter.value_() / 1_000_000