summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--MAINTAINERS3
-rw-r--r--block/bfq-iosched.c68
-rw-r--r--drivers/ata/libahci_platform.c3
-rw-r--r--drivers/ata/libata-scsi.c21
-rw-r--r--drivers/ata/libata-sff.c6
-rw-r--r--drivers/ata/pata_rb532_cf.c1
-rw-r--r--drivers/block/aoe/aoedev.c13
-rw-r--r--drivers/block/loop.c18
-rw-r--r--drivers/block/nbd.c2
-rw-r--r--drivers/md/bcache/sysfs.c20
-rw-r--r--drivers/s390/block/dasd_alias.c22
-rw-r--r--fs/block_dev.c113
-rw-r--r--fs/io_uring.c3
-rw-r--r--include/linux/fs.h6
14 files changed, 209 insertions, 90 deletions
diff --git a/MAINTAINERS b/MAINTAINERS
index 6426db5198f0..6c49b48cfd69 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6322,7 +6322,8 @@ F: Documentation/devicetree/bindings/counter/ftm-quaddec.txt
F: drivers/counter/ftm-quaddec.c
FLOPPY DRIVER
-S: Orphan
+M: Denis Efremov <efremov@linux.com>
+S: Odd Fixes
L: linux-block@vger.kernel.org
F: drivers/block/floppy.c
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 586fcfe227ea..b33be928d164 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -1924,12 +1924,13 @@ static void bfq_add_request(struct request *rq)
* confirmed no later than during the next
* I/O-plugging interval for bfqq.
*/
- if (!bfq_bfqq_has_short_ttime(bfqq) &&
+ if (bfqd->last_completed_rq_bfqq &&
+ !bfq_bfqq_has_short_ttime(bfqq) &&
ktime_get_ns() - bfqd->last_completion <
200 * NSEC_PER_USEC) {
if (bfqd->last_completed_rq_bfqq != bfqq &&
- bfqd->last_completed_rq_bfqq !=
- bfqq->waker_bfqq) {
+ bfqd->last_completed_rq_bfqq !=
+ bfqq->waker_bfqq) {
/*
* First synchronization detected with
* a candidate waker queue, or with a
@@ -2250,9 +2251,14 @@ static void bfq_request_merged(struct request_queue *q, struct request *req,
blk_rq_pos(container_of(rb_prev(&req->rb_node),
struct request, rb_node))) {
struct bfq_queue *bfqq = bfq_init_rq(req);
- struct bfq_data *bfqd = bfqq->bfqd;
+ struct bfq_data *bfqd;
struct request *prev, *next_rq;
+ if (!bfqq)
+ return;
+
+ bfqd = bfqq->bfqd;
+
/* Reposition request in its sort_list */
elv_rb_del(&bfqq->sort_list, req);
elv_rb_add(&bfqq->sort_list, req);
@@ -2299,6 +2305,9 @@ static void bfq_requests_merged(struct request_queue *q, struct request *rq,
struct bfq_queue *bfqq = bfq_init_rq(rq),
*next_bfqq = bfq_init_rq(next);
+ if (!bfqq)
+ return;
+
/*
* If next and rq belong to the same bfq_queue and next is older
* than rq, then reposition rq in the fifo (by substituting next
@@ -4764,6 +4773,8 @@ static struct request *bfq_dispatch_request(struct blk_mq_hw_ctx *hctx)
*/
void bfq_put_queue(struct bfq_queue *bfqq)
{
+ struct bfq_queue *item;
+ struct hlist_node *n;
#ifdef CONFIG_BFQ_GROUP_IOSCHED
struct bfq_group *bfqg = bfqq_group(bfqq);
#endif
@@ -4808,6 +4819,36 @@ void bfq_put_queue(struct bfq_queue *bfqq)
bfqq->bfqd->burst_size--;
}
+ /*
+ * bfqq does not exist any longer, so it cannot be woken by
+ * any other queue, and cannot wake any other queue. Then bfqq
+ * must be removed from the woken list of its possible waker
+ * queue, and all queues in the woken list of bfqq must stop
+ * having a waker queue. Strictly speaking, these updates
+ * should be performed when bfqq remains with no I/O source
+ * attached to it, which happens before bfqq gets freed. In
+ * particular, this happens when the last process associated
+ * with bfqq exits or gets associated with a different
+ * queue. However, both events lead to bfqq being freed soon,
+ * and dangling references would come out only after bfqq gets
+ * freed. So these updates are done here, as a simple and safe
+ * way to handle all cases.
+ */
+ /* remove bfqq from woken list */
+ if (!hlist_unhashed(&bfqq->woken_list_node))
+ hlist_del_init(&bfqq->woken_list_node);
+
+ /* reset waker for all queues in woken list */
+ hlist_for_each_entry_safe(item, n, &bfqq->woken_list,
+ woken_list_node) {
+ item->waker_bfqq = NULL;
+ bfq_clear_bfqq_has_waker(item);
+ hlist_del_init(&item->woken_list_node);
+ }
+
+ if (bfqq->bfqd && bfqq->bfqd->last_completed_rq_bfqq == bfqq)
+ bfqq->bfqd->last_completed_rq_bfqq = NULL;
+
kmem_cache_free(bfq_pool, bfqq);
#ifdef CONFIG_BFQ_GROUP_IOSCHED
bfqg_and_blkg_put(bfqg);
@@ -4835,9 +4876,6 @@ static void bfq_put_cooperator(struct bfq_queue *bfqq)
static void bfq_exit_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq)
{
- struct bfq_queue *item;
- struct hlist_node *n;
-
if (bfqq == bfqd->in_service_queue) {
__bfq_bfqq_expire(bfqd, bfqq, BFQQE_BUDGET_TIMEOUT);
bfq_schedule_dispatch(bfqd);
@@ -4847,18 +4885,6 @@ static void bfq_exit_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq)
bfq_put_cooperator(bfqq);
- /* remove bfqq from woken list */
- if (!hlist_unhashed(&bfqq->woken_list_node))
- hlist_del_init(&bfqq->woken_list_node);
-
- /* reset waker for all queues in woken list */
- hlist_for_each_entry_safe(item, n, &bfqq->woken_list,
- woken_list_node) {
- item->waker_bfqq = NULL;
- bfq_clear_bfqq_has_waker(item);
- hlist_del_init(&item->woken_list_node);
- }
-
bfq_put_queue(bfqq); /* release process reference */
}
@@ -5436,12 +5462,12 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
spin_lock_irq(&bfqd->lock);
bfqq = bfq_init_rq(rq);
- if (at_head || blk_rq_is_passthrough(rq)) {
+ if (!bfqq || at_head || blk_rq_is_passthrough(rq)) {
if (at_head)
list_add(&rq->queuelist, &bfqd->dispatch);
else
list_add_tail(&rq->queuelist, &bfqd->dispatch);
- } else { /* bfqq is assumed to be non null here */
+ } else {
idle_timer_disabled = __bfq_insert_request(bfqd, rq);
/*
* Update bfqq, because, if a queue merge has occurred
diff --git a/drivers/ata/libahci_platform.c b/drivers/ata/libahci_platform.c
index 3a36e76eca83..9e9583a6bba9 100644
--- a/drivers/ata/libahci_platform.c
+++ b/drivers/ata/libahci_platform.c
@@ -338,6 +338,9 @@ static int ahci_platform_get_phy(struct ahci_host_priv *hpriv, u32 port,
hpriv->phys[port] = NULL;
rc = 0;
break;
+ case -EPROBE_DEFER:
+ /* Do not complain yet */
+ break;
default:
dev_err(dev,
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 391ac0503dc0..76d0f9de767b 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -1786,6 +1786,21 @@ nothing_to_do:
return 1;
}
+static bool ata_check_nblocks(struct scsi_cmnd *scmd, u32 n_blocks)
+{
+ struct request *rq = scmd->request;
+ u32 req_blocks;
+
+ if (!blk_rq_is_passthrough(rq))
+ return true;
+
+ req_blocks = blk_rq_bytes(rq) / scmd->device->sector_size;
+ if (n_blocks > req_blocks)
+ return false;
+
+ return true;
+}
+
/**
* ata_scsi_rw_xlat - Translate SCSI r/w command into an ATA one
* @qc: Storage for translated ATA taskfile
@@ -1830,6 +1845,8 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc)
scsi_10_lba_len(cdb, &block, &n_block);
if (cdb[1] & (1 << 3))
tf_flags |= ATA_TFLAG_FUA;
+ if (!ata_check_nblocks(scmd, n_block))
+ goto invalid_fld;
break;
case READ_6:
case WRITE_6:
@@ -1844,6 +1861,8 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc)
*/
if (!n_block)
n_block = 256;
+ if (!ata_check_nblocks(scmd, n_block))
+ goto invalid_fld;
break;
case READ_16:
case WRITE_16:
@@ -1854,6 +1873,8 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc)
scsi_16_lba_len(cdb, &block, &n_block);
if (cdb[1] & (1 << 3))
tf_flags |= ATA_TFLAG_FUA;
+ if (!ata_check_nblocks(scmd, n_block))
+ goto invalid_fld;
break;
default:
DPRINTK("no-byte command\n");
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 10aa27882142..4f115adb4ee8 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -658,6 +658,10 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
unsigned int offset;
unsigned char *buf;
+ if (!qc->cursg) {
+ qc->curbytes = qc->nbytes;
+ return;
+ }
if (qc->curbytes == qc->nbytes - qc->sect_size)
ap->hsm_task_state = HSM_ST_LAST;
@@ -683,6 +687,8 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
if (qc->cursg_ofs == qc->cursg->length) {
qc->cursg = sg_next(qc->cursg);
+ if (!qc->cursg)
+ ap->hsm_task_state = HSM_ST_LAST;
qc->cursg_ofs = 0;
}
}
diff --git a/drivers/ata/pata_rb532_cf.c b/drivers/ata/pata_rb532_cf.c
index 7c37f2ff09e4..deae466395de 100644
--- a/drivers/ata/pata_rb532_cf.c
+++ b/drivers/ata/pata_rb532_cf.c
@@ -158,7 +158,6 @@ static int rb532_pata_driver_probe(struct platform_device *pdev)
static int rb532_pata_driver_remove(struct platform_device *pdev)
{
struct ata_host *ah = platform_get_drvdata(pdev);
- struct rb532_cf_info *info = ah->private_data;
ata_host_detach(ah);
diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c
index 5b49f1b33ebe..e2ea2356da06 100644
--- a/drivers/block/aoe/aoedev.c
+++ b/drivers/block/aoe/aoedev.c
@@ -323,10 +323,14 @@ flush(const char __user *str, size_t cnt, int exiting)
}
flush_scheduled_work();
- /* pass one: without sleeping, do aoedev_downdev */
+ /* pass one: do aoedev_downdev, which might sleep */
+restart1:
spin_lock_irqsave(&devlist_lock, flags);
for (d = devlist; d; d = d->next) {
spin_lock(&d->lock);
+ if (d->flags & DEVFL_TKILL)
+ goto cont;
+
if (exiting) {
/* unconditionally take each device down */
} else if (specified) {
@@ -338,8 +342,11 @@ flush(const char __user *str, size_t cnt, int exiting)
|| d->ref)
goto cont;
+ spin_unlock(&d->lock);
+ spin_unlock_irqrestore(&devlist_lock, flags);
aoedev_downdev(d);
d->flags |= DEVFL_TKILL;
+ goto restart1;
cont:
spin_unlock(&d->lock);
}
@@ -348,7 +355,7 @@ cont:
/* pass two: call freedev, which might sleep,
* for aoedevs marked with DEVFL_TKILL
*/
-restart:
+restart2:
spin_lock_irqsave(&devlist_lock, flags);
for (d = devlist; d; d = d->next) {
spin_lock(&d->lock);
@@ -357,7 +364,7 @@ restart:
spin_unlock(&d->lock);
spin_unlock_irqrestore(&devlist_lock, flags);
freedev(d);
- goto restart;
+ goto restart2;
}
spin_unlock(&d->lock);
}
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 44c9985f352a..ab7ca5989097 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -885,7 +885,7 @@ static void loop_unprepare_queue(struct loop_device *lo)
static int loop_kthread_worker_fn(void *worker_ptr)
{
- current->flags |= PF_LESS_THROTTLE;
+ current->flags |= PF_LESS_THROTTLE | PF_MEMALLOC_NOIO;
return kthread_worker_fn(worker_ptr);
}
@@ -924,6 +924,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
struct file *file;
struct inode *inode;
struct address_space *mapping;
+ struct block_device *claimed_bdev = NULL;
int lo_flags = 0;
int error;
loff_t size;
@@ -942,10 +943,11 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
* here to avoid changing device under exclusive owner.
*/
if (!(mode & FMODE_EXCL)) {
- bdgrab(bdev);
- error = blkdev_get(bdev, mode | FMODE_EXCL, loop_set_fd);
- if (error)
+ claimed_bdev = bd_start_claiming(bdev, loop_set_fd);
+ if (IS_ERR(claimed_bdev)) {
+ error = PTR_ERR(claimed_bdev);
goto out_putf;
+ }
}
error = mutex_lock_killable(&loop_ctl_mutex);
@@ -1015,15 +1017,15 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
mutex_unlock(&loop_ctl_mutex);
if (partscan)
loop_reread_partitions(lo, bdev);
- if (!(mode & FMODE_EXCL))
- blkdev_put(bdev, mode | FMODE_EXCL);
+ if (claimed_bdev)
+ bd_abort_claiming(bdev, claimed_bdev, loop_set_fd);
return 0;
out_unlock:
mutex_unlock(&loop_ctl_mutex);
out_bdev:
- if (!(mode & FMODE_EXCL))
- blkdev_put(bdev, mode | FMODE_EXCL);
+ if (claimed_bdev)
+ bd_abort_claiming(bdev, claimed_bdev, loop_set_fd);
out_putf:
fput(file);
out:
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 9bcde2325893..e21d2ded732b 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -1231,7 +1231,7 @@ static void nbd_clear_sock_ioctl(struct nbd_device *nbd,
struct block_device *bdev)
{
sock_shutdown(nbd);
- kill_bdev(bdev);
+ __invalidate_device(bdev, true);
nbd_bdev_reset(bdev);
if (test_and_clear_bit(NBD_HAS_CONFIG_REF,
&nbd->config->runtime_flags))
diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c
index 9f0826712845..e2059af90791 100644
--- a/drivers/md/bcache/sysfs.c
+++ b/drivers/md/bcache/sysfs.c
@@ -23,24 +23,28 @@ static const char * const bch_cache_modes[] = {
"writethrough",
"writeback",
"writearound",
- "none"
+ "none",
+ NULL
};
/* Default is 0 ("auto") */
static const char * const bch_stop_on_failure_modes[] = {
"auto",
- "always"
+ "always",
+ NULL
};
static const char * const cache_replacement_policies[] = {
"lru",
"fifo",
- "random"
+ "random",
+ NULL
};
static const char * const error_actions[] = {
"unregister",
- "panic"
+ "panic",
+ NULL
};
write_attribute(attach);
@@ -338,7 +342,7 @@ STORE(__cached_dev)
}
if (attr == &sysfs_cache_mode) {
- v = sysfs_match_string(bch_cache_modes, buf);
+ v = __sysfs_match_string(bch_cache_modes, -1, buf);
if (v < 0)
return v;
@@ -349,7 +353,7 @@ STORE(__cached_dev)
}
if (attr == &sysfs_stop_when_cache_set_failed) {
- v = sysfs_match_string(bch_stop_on_failure_modes, buf);
+ v = __sysfs_match_string(bch_stop_on_failure_modes, -1, buf);
if (v < 0)
return v;
@@ -816,7 +820,7 @@ STORE(__bch_cache_set)
0, UINT_MAX);
if (attr == &sysfs_errors) {
- v = sysfs_match_string(error_actions, buf);
+ v = __sysfs_match_string(error_actions, -1, buf);
if (v < 0)
return v;
@@ -1088,7 +1092,7 @@ STORE(__bch_cache)
}
if (attr == &sysfs_cache_replacement_policy) {
- v = sysfs_match_string(cache_replacement_policies, buf);
+ v = __sysfs_match_string(cache_replacement_policies, -1, buf);
if (v < 0)
return v;
diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c
index b9ce93e9df89..99f86612f775 100644
--- a/drivers/s390/block/dasd_alias.c
+++ b/drivers/s390/block/dasd_alias.c
@@ -383,6 +383,20 @@ suborder_not_supported(struct dasd_ccw_req *cqr)
char msg_format;
char msg_no;
+ /*
+ * intrc values ENODEV, ENOLINK and EPERM
+ * will be optained from sleep_on to indicate that no
+ * IO operation can be started
+ */
+ if (cqr->intrc == -ENODEV)
+ return 1;
+
+ if (cqr->intrc == -ENOLINK)
+ return 1;
+
+ if (cqr->intrc == -EPERM)
+ return 1;
+
sense = dasd_get_sense(&cqr->irb);
if (!sense)
return 0;
@@ -447,12 +461,8 @@ static int read_unit_address_configuration(struct dasd_device *device,
lcu->flags &= ~NEED_UAC_UPDATE;
spin_unlock_irqrestore(&lcu->lock, flags);
- do {
- rc = dasd_sleep_on(cqr);
- if (rc && suborder_not_supported(cqr))
- return -EOPNOTSUPP;
- } while (rc && (cqr->retries > 0));
- if (rc) {
+ rc = dasd_sleep_on(cqr);
+ if (rc && !suborder_not_supported(cqr)) {
spin_lock_irqsave(&lcu->lock, flags);
lcu->flags |= NEED_UAC_UPDATE;
spin_unlock_irqrestore(&lcu->lock, flags);
diff --git a/fs/block_dev.c b/fs/block_dev.c
index c2a85b587922..eb657ab94060 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -349,7 +349,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
loff_t pos = iocb->ki_pos;
blk_qc_t qc = BLK_QC_T_NONE;
gfp_t gfp;
- ssize_t ret;
+ int ret;
if ((pos | iov_iter_alignment(iter)) &
(bdev_logical_block_size(bdev) - 1))
@@ -386,8 +386,6 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
ret = 0;
for (;;) {
- int err;
-
bio_set_dev(bio, bdev);
bio->bi_iter.bi_sector = pos >> 9;
bio->bi_write_hint = iocb->ki_hint;
@@ -395,10 +393,8 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
bio->bi_end_io = blkdev_bio_end_io;
bio->bi_ioprio = iocb->ki_ioprio;
- err = bio_iov_iter_get_pages(bio, iter);
- if (unlikely(err)) {
- if (!ret)
- ret = err;
+ ret = bio_iov_iter_get_pages(bio, iter);
+ if (unlikely(ret)) {
bio->bi_status = BLK_STS_IOERR;
bio_endio(bio);
break;
@@ -421,7 +417,6 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
if (nowait)
bio->bi_opf |= (REQ_NOWAIT | REQ_NOWAIT_INLINE);
- dio->size += bio->bi_iter.bi_size;
pos += bio->bi_iter.bi_size;
nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES);
@@ -433,10 +428,11 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
polled = true;
}
+ dio->size += bio->bi_iter.bi_size;
qc = submit_bio(bio);
if (qc == BLK_QC_T_EAGAIN) {
- if (!ret)
- ret = -EAGAIN;
+ dio->size -= bio->bi_iter.bi_size;
+ ret = -EAGAIN;
goto error;
}
@@ -459,18 +455,17 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
atomic_inc(&dio->ref);
}
+ dio->size += bio->bi_iter.bi_size;
qc = submit_bio(bio);
if (qc == BLK_QC_T_EAGAIN) {
- if (!ret)
- ret = -EAGAIN;
+ dio->size -= bio->bi_iter.bi_size;
+ ret = -EAGAIN;
goto error;
}
- ret += bio->bi_iter.bi_size;
bio = bio_alloc(gfp, nr_pages);
if (!bio) {
- if (!ret)
- ret = -EAGAIN;
+ ret = -EAGAIN;
goto error;
}
}
@@ -495,6 +490,8 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
out:
if (!ret)
ret = blk_status_to_errno(dio->bio.bi_status);
+ if (likely(!ret))
+ ret = dio->size;
bio_put(&dio->bio);
return ret;
@@ -1181,8 +1178,7 @@ static struct gendisk *bdev_get_gendisk(struct block_device *bdev, int *partno)
* Pointer to the block device containing @bdev on success, ERR_PTR()
* value on failure.
*/
-static struct block_device *bd_start_claiming(struct block_device *bdev,
- void *holder)
+struct block_device *bd_start_claiming(struct block_device *bdev, void *holder)
{
struct gendisk *disk;
struct block_device *whole;
@@ -1229,6 +1225,62 @@ static struct block_device *bd_start_claiming(struct block_device *bdev,
return ERR_PTR(err);
}
}
+EXPORT_SYMBOL(bd_start_claiming);
+
+static void bd_clear_claiming(struct block_device *whole, void *holder)
+{
+ lockdep_assert_held(&bdev_lock);
+ /* tell others that we're done */
+ BUG_ON(whole->bd_claiming != holder);
+ whole->bd_claiming = NULL;
+ wake_up_bit(&whole->bd_claiming, 0);
+}
+
+/**
+ * bd_finish_claiming - finish claiming of a block device
+ * @bdev: block device of interest
+ * @whole: whole block device (returned from bd_start_claiming())
+ * @holder: holder that has claimed @bdev
+ *
+ * Finish exclusive open of a block device. Mark the device as exlusively
+ * open by the holder and wake up all waiters for exclusive open to finish.
+ */
+void bd_finish_claiming(struct block_device *bdev, struct block_device *whole,
+ void *holder)
+{
+ spin_lock(&bdev_lock);
+ BUG_ON(!bd_may_claim(bdev, whole, holder));
+ /*
+ * Note that for a whole device bd_holders will be incremented twice,
+ * and bd_holder will be set to bd_may_claim before being set to holder
+ */
+ whole->bd_holders++;
+ whole->bd_holder = bd_may_claim;
+ bdev->bd_holders++;
+ bdev->bd_holder = holder;
+ bd_clear_claiming(whole, holder);
+ spin_unlock(&bdev_lock);
+}
+EXPORT_SYMBOL(bd_finish_claiming);
+
+/**
+ * bd_abort_claiming - abort claiming of a block device
+ * @bdev: block device of interest
+ * @whole: whole block device (returned from bd_start_claiming())
+ * @holder: holder that has claimed @bdev
+ *
+ * Abort claiming of a block device when the exclusive open failed. This can be
+ * also used when exclusive open is not actually desired and we just needed
+ * to block other exclusive openers for a while.
+ */
+void bd_abort_claiming(struct block_device *bdev, struct block_device *whole,
+ void *holder)
+{
+ spin_lock(&bdev_lock);
+ bd_clear_claiming(whole, holder);
+ spin_unlock(&bdev_lock);
+}
+EXPORT_SYMBOL(bd_abort_claiming);
#ifdef CONFIG_SYSFS
struct bd_holder_disk {
@@ -1698,29 +1750,10 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder)
/* finish claiming */
mutex_lock(&bdev->bd_mutex);
- spin_lock(&bdev_lock);
-
- if (!res) {
- BUG_ON(!bd_may_claim(bdev, whole, holder));
- /*
- * Note that for a whole device bd_holders
- * will be incremented twice, and bd_holder
- * will be set to bd_may_claim before being
- * set to holder
- */
- whole->bd_holders++;
- whole->bd_holder = bd_may_claim;
- bdev->bd_holders++;
- bdev->bd_holder = holder;
- }
-
- /* tell others that we're done */
- BUG_ON(whole->bd_claiming != holder);
- whole->bd_claiming = NULL;
- wake_up_bit(&whole->bd_claiming, 0);
-
- spin_unlock(&bdev_lock);
-
+ if (!res)
+ bd_finish_claiming(bdev, whole, holder);
+ else
+ bd_abort_claiming(bdev, whole, holder);
/*
* Block event polling for write claims if requested. Any
* write holder makes the write_holder state stick until
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 012bc0efb9d3..d542f1cf4428 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1838,6 +1838,7 @@ restart:
do {
struct sqe_submit *s = &req->submit;
const struct io_uring_sqe *sqe = s->sqe;
+ unsigned int flags = req->flags;
/* Ensure we clear previously set non-block flag */
req->rw.ki_flags &= ~IOCB_NOWAIT;
@@ -1883,7 +1884,7 @@ restart:
kfree(sqe);
/* req from defer and link list needn't decrease async cnt */
- if (req->flags & (REQ_F_IO_DRAINED | REQ_F_LINK_DONE))
+ if (flags & (REQ_F_IO_DRAINED | REQ_F_LINK_DONE))
goto out;
if (!async_list)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 56b8e358af5c..997a530ff4e9 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2598,6 +2598,12 @@ extern struct block_device *blkdev_get_by_path(const char *path, fmode_t mode,
void *holder);
extern struct block_device *blkdev_get_by_dev(dev_t dev, fmode_t mode,
void *holder);
+extern struct block_device *bd_start_claiming(struct block_device *bdev,
+ void *holder);
+extern void bd_finish_claiming(struct block_device *bdev,
+ struct block_device *whole, void *holder);
+extern void bd_abort_claiming(struct block_device *bdev,
+ struct block_device *whole, void *holder);
extern void blkdev_put(struct block_device *bdev, fmode_t mode);
extern int __blkdev_reread_part(struct block_device *bdev);
extern int blkdev_reread_part(struct block_device *bdev);
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-13 14:08:49 +0000