Merge tag 'block-6.17-20250815' of git://git.kernel.dk/linux

Pull block fixes from Jens Axboe:

 - Fix for unprivileged daemons in ublk

 - Speedup ublk release by removing unnecessary quiesce

 - Fix for blk-wbt, where a regression caused it to not be possible to
   enable at runtime

 - blk-wbt cleanups

 - Kill the page pool from drbd

 - Remove redundant __GFP_NOWARN uses in a few spots

 - Fix for a kobject double initialization issues

* tag 'block-6.17-20250815' of git://git.kernel.dk/linux:
  block: restore default wbt enablement
  Docs: admin-guide: Correct spelling mistake
  blk-wbt: doc: Update the doc of the wbt_lat_usec interface
  blk-wbt: Eliminate ambiguity in the comments of struct rq_wb
  blk-wbt: Optimize wbt_done() for non-throttled writes
  block: fix kobject double initialization in add_disk
  blk-cgroup: remove redundant __GFP_NOWARN
  block, bfq: remove redundant __GFP_NOWARN
  ublk: check for unprivileged daemon on each I/O fetch
  ublk: don't quiesce in ublk_ch_release
  drbd: Remove the open-coded page pool

Author: torvalds

Documentation/ABI/stable/sysfs-block

@@ -731,7 +731,7 @@ Contact:	[email protected]
 Description:
 		[RW] If the device is registered for writeback throttling, then
 		this file shows the target minimum read latency. If this latency
-		is exceeded in a given window of time (see wb_window_usec), then
+		is exceeded in a given window of time (see curr_win_nsec), then
 		the writeback throttling will start scaling back writes. Writing
 		a value of '0' to this file disables the feature. Writing a
 		value of '-1' to this file resets the value to the default

Documentation/admin-guide/blockdev/zoned_loop.rst

@@ -79,7 +79,7 @@ zone_capacity_mb   Device zone capacity (must always be equal to or lower than
                    the zone size. Default: zone size.
 conv_zones         Total number of conventioanl zones starting from sector 0.
                    Default: 8.
-base_dir           Path to the base directoy where to create the directory
+base_dir           Path to the base directory where to create the directory
                    containing the zone files of the device.
                    Default=/var/local/zloop.
                    The device directory containing the zone files is always

block/bfq-iosched.c

@@ -5847,8 +5847,7 @@ static struct bfq_queue *bfq_get_queue(struct bfq_data *bfqd,
 			goto out;
 	}
 
-	bfqq = kmem_cache_alloc_node(bfq_pool,
-				     GFP_NOWAIT | __GFP_ZERO | __GFP_NOWARN,
+	bfqq = kmem_cache_alloc_node(bfq_pool, GFP_NOWAIT | __GFP_ZERO,
 				     bfqd->queue->node);
 
 	if (bfqq) {

block/blk-cgroup.c

@@ -394,7 +394,7 @@ static struct blkcg_gq *blkg_create(struct blkcg *blkcg, struct gendisk *disk,
 
 	/* allocate */
 	if (!new_blkg) {
-		new_blkg = blkg_alloc(blkcg, disk, GFP_NOWAIT | __GFP_NOWARN);
+		new_blkg = blkg_alloc(blkcg, disk, GFP_NOWAIT);
 		if (unlikely(!new_blkg)) {
 			ret = -ENOMEM;
 			goto err_put_css;
@@ -1467,7 +1467,7 @@ blkcg_css_alloc(struct cgroup_subsys_state *parent_css)
 
 	spin_lock_init(&blkcg->lock);
 	refcount_set(&blkcg->online_pin, 1);
-	INIT_RADIX_TREE(&blkcg->blkg_tree, GFP_NOWAIT | __GFP_NOWARN);
+	INIT_RADIX_TREE(&blkcg->blkg_tree, GFP_NOWAIT);
 	INIT_HLIST_HEAD(&blkcg->blkg_list);
 #ifdef CONFIG_CGROUP_WRITEBACK
 	INIT_LIST_HEAD(&blkcg->cgwb_list);
@@ -1630,7 +1630,7 @@ int blkcg_activate_policy(struct gendisk *disk, const struct blkcg_policy *pol)
 			pd_prealloc = NULL;
 		} else {
 			pd = pol->pd_alloc_fn(disk, blkg->blkcg,
-					      GFP_NOWAIT | __GFP_NOWARN);
+					      GFP_NOWAIT);
 		}
 
 		if (!pd) {

block/blk-sysfs.c

@@ -847,7 +847,7 @@ static void blk_queue_release(struct kobject *kobj)
 	/* nothing to do here, all data is associated with the parent gendisk */
 }
 
-static const struct kobj_type blk_queue_ktype = {
+const struct kobj_type blk_queue_ktype = {
 	.default_groups = blk_queue_attr_groups,
 	.sysfs_ops	= &queue_sysfs_ops,
 	.release	= blk_queue_release,
@@ -875,15 +875,14 @@ int blk_register_queue(struct gendisk *disk)
 	struct request_queue *q = disk->queue;
 	int ret;
 
-	kobject_init(&disk->queue_kobj, &blk_queue_ktype);
 	ret = kobject_add(&disk->queue_kobj, &disk_to_dev(disk)->kobj, "queue");
 	if (ret < 0)
-		goto out_put_queue_kobj;
+		return ret;
 
 	if (queue_is_mq(q)) {
 		ret = blk_mq_sysfs_register(disk);
 		if (ret)
-			goto out_put_queue_kobj;
+			goto out_del_queue_kobj;
 	}
 	mutex_lock(&q->sysfs_lock);
 
@@ -903,9 +902,9 @@ int blk_register_queue(struct gendisk *disk)
 
 	if (queue_is_mq(q))
 		elevator_set_default(q);
-	wbt_enable_default(disk);
 
 	blk_queue_flag_set(QUEUE_FLAG_REGISTERED, q);
+	wbt_enable_default(disk);
 
 	/* Now everything is ready and send out KOBJ_ADD uevent */
 	kobject_uevent(&disk->queue_kobj, KOBJ_ADD);
@@ -934,8 +933,8 @@ int blk_register_queue(struct gendisk *disk)
 	mutex_unlock(&q->sysfs_lock);
 	if (queue_is_mq(q))
 		blk_mq_sysfs_unregister(disk);
-out_put_queue_kobj:
-	kobject_put(&disk->queue_kobj);
+out_del_queue_kobj:
+	kobject_del(&disk->queue_kobj);
 	return ret;
 }
 
@@ -986,5 +985,4 @@ void blk_unregister_queue(struct gendisk *disk)
 		elevator_set_none(q);
 
 	blk_debugfs_remove(disk);
-	kobject_put(&disk->queue_kobj);
 }

block/blk-wbt.c

@@ -85,8 +85,8 @@ struct rq_wb {
 	u64 sync_issue;
 	void *sync_cookie;
 
-	unsigned long last_issue;		/* last non-throttled issue */
-	unsigned long last_comp;		/* last non-throttled comp */
+	unsigned long last_issue;	/* issue time of last read rq */
+	unsigned long last_comp;	/* completion time of last read rq */
 	unsigned long min_lat_nsec;
 	struct rq_qos rqos;
 	struct rq_wait rq_wait[WBT_NUM_RWQ];
@@ -248,13 +248,14 @@ static void wbt_done(struct rq_qos *rqos, struct request *rq)
 	struct rq_wb *rwb = RQWB(rqos);
 
 	if (!wbt_is_tracked(rq)) {
-		if (rwb->sync_cookie == rq) {
-			rwb->sync_issue = 0;
-			rwb->sync_cookie = NULL;
-		}
+		if (wbt_is_read(rq)) {
+			if (rwb->sync_cookie == rq) {
+				rwb->sync_issue = 0;
+				rwb->sync_cookie = NULL;
+			}
 
-		if (wbt_is_read(rq))
 			wb_timestamp(rwb, &rwb->last_comp);
+		}
 	} else {
 		WARN_ON_ONCE(rq == rwb->sync_cookie);
 		__wbt_done(rqos, wbt_flags(rq));

block/blk.h

@@ -29,6 +29,7 @@ struct elevator_tags;
 /* Max future timer expiry for timeouts */
 #define BLK_MAX_TIMEOUT		(5 * HZ)
 
+extern const struct kobj_type blk_queue_ktype;
 extern struct dentry *blk_debugfs_root;
 
 struct blk_flush_queue {

block/genhd.c

@@ -1303,6 +1303,7 @@ static void disk_release(struct device *dev)
 	disk_free_zone_resources(disk);
 	xa_destroy(&disk->part_tbl);
 
+	kobject_put(&disk->queue_kobj);
 	disk->queue->disk = NULL;
 	blk_put_queue(disk->queue);
 
@@ -1486,6 +1487,7 @@ struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id,
 	INIT_LIST_HEAD(&disk->slave_bdevs);
 #endif
 	mutex_init(&disk->rqos_state_mutex);
+	kobject_init(&disk->queue_kobj, &blk_queue_ktype);
 	return disk;
 
 out_erase_part0:

drivers/block/drbd/drbd_int.h

@@ -380,6 +380,9 @@ enum {
 	/* this is/was a write request */
 	__EE_WRITE,
 
+	/* hand back using mempool_free(e, drbd_buffer_page_pool) */
+	__EE_RELEASE_TO_MEMPOOL,
+
 	/* this is/was a write same request */
 	__EE_WRITE_SAME,
 
@@ -402,6 +405,7 @@ enum {
 #define EE_IN_INTERVAL_TREE	(1<<__EE_IN_INTERVAL_TREE)
 #define EE_SUBMITTED		(1<<__EE_SUBMITTED)
 #define EE_WRITE		(1<<__EE_WRITE)
+#define EE_RELEASE_TO_MEMPOOL	(1<<__EE_RELEASE_TO_MEMPOOL)
 #define EE_WRITE_SAME		(1<<__EE_WRITE_SAME)
 #define EE_APPLICATION		(1<<__EE_APPLICATION)
 #define EE_RS_THIN_REQ		(1<<__EE_RS_THIN_REQ)
@@ -858,7 +862,6 @@ struct drbd_device {
 	struct list_head sync_ee;   /* IO in progress (P_RS_DATA_REPLY gets written to disk) */
 	struct list_head done_ee;   /* need to send P_WRITE_ACK */
 	struct list_head read_ee;   /* [RS]P_DATA_REQUEST being read */
-	struct list_head net_ee;    /* zero-copy network send in progress */
 
 	struct list_head resync_reads;
 	atomic_t pp_in_use;		/* allocated from page pool */
@@ -1329,31 +1332,14 @@ extern struct kmem_cache *drbd_al_ext_cache;	/* activity log extents */
 extern mempool_t drbd_request_mempool;
 extern mempool_t drbd_ee_mempool;
 
-/* drbd's page pool, used to buffer data received from the peer,
- * or data requested by the peer.
- *
- * This does not have an emergency reserve.
- *
- * When allocating from this pool, it first takes pages from the pool.
- * Only if the pool is depleted will try to allocate from the system.
- *
- * The assumption is that pages taken from this pool will be processed,
- * and given back, "quickly", and then can be recycled, so we can avoid
- * frequent calls to alloc_page(), and still will be able to make progress even
- * under memory pressure.
- */
-extern struct page *drbd_pp_pool;
-extern spinlock_t   drbd_pp_lock;
-extern int	    drbd_pp_vacant;
-extern wait_queue_head_t drbd_pp_wait;
-
 /* We also need a standard (emergency-reserve backed) page pool
  * for meta data IO (activity log, bitmap).
  * We can keep it global, as long as it is used as "N pages at a time".
  * 128 should be plenty, currently we probably can get away with as few as 1.
  */
 #define DRBD_MIN_POOL_PAGES	128
 extern mempool_t drbd_md_io_page_pool;
+extern mempool_t drbd_buffer_page_pool;
 
 /* We also need to make sure we get a bio
  * when we need it for housekeeping purposes */
@@ -1488,10 +1474,7 @@ extern struct drbd_peer_request *drbd_alloc_peer_req(struct drbd_peer_device *,
 						     sector_t, unsigned int,
 						     unsigned int,
 						     gfp_t) __must_hold(local);
-extern void __drbd_free_peer_req(struct drbd_device *, struct drbd_peer_request *,
-				 int);
-#define drbd_free_peer_req(m,e) __drbd_free_peer_req(m, e, 0)
-#define drbd_free_net_peer_req(m,e) __drbd_free_peer_req(m, e, 1)
+extern void drbd_free_peer_req(struct drbd_device *device, struct drbd_peer_request *req);
 extern struct page *drbd_alloc_pages(struct drbd_peer_device *, unsigned int, bool);
 extern void _drbd_clear_done_ee(struct drbd_device *device, struct list_head *to_be_freed);
 extern int drbd_connected(struct drbd_peer_device *);
@@ -1610,16 +1593,6 @@ static inline struct page *page_chain_next(struct page *page)
 	for (; page && ({ n = page_chain_next(page); 1; }); page = n)
 
 
-static inline int drbd_peer_req_has_active_page(struct drbd_peer_request *peer_req)
-{
-	struct page *page = peer_req->pages;
-	page_chain_for_each(page) {
-		if (page_count(page) > 1)
-			return 1;
-	}
-	return 0;
-}
-
 static inline union drbd_state drbd_read_state(struct drbd_device *device)
 {
 	struct drbd_resource *resource = device->resource;

drivers/block/drbd/drbd_main.c

@@ -114,20 +114,10 @@ struct kmem_cache *drbd_al_ext_cache;	/* activity log extents */
 mempool_t drbd_request_mempool;
 mempool_t drbd_ee_mempool;
 mempool_t drbd_md_io_page_pool;
+mempool_t drbd_buffer_page_pool;
 struct bio_set drbd_md_io_bio_set;
 struct bio_set drbd_io_bio_set;
 
-/* I do not use a standard mempool, because:
-   1) I want to hand out the pre-allocated objects first.
-   2) I want to be able to interrupt sleeping allocation with a signal.
-   Note: This is a single linked list, the next pointer is the private
-	 member of struct page.
- */
-struct page *drbd_pp_pool;
-DEFINE_SPINLOCK(drbd_pp_lock);
-int          drbd_pp_vacant;
-wait_queue_head_t drbd_pp_wait;
-
 DEFINE_RATELIMIT_STATE(drbd_ratelimit_state, 5 * HZ, 5);
 
 static const struct block_device_operations drbd_ops = {
@@ -1611,6 +1601,7 @@ static int _drbd_send_zc_bio(struct drbd_peer_device *peer_device, struct bio *b
 static int _drbd_send_zc_ee(struct drbd_peer_device *peer_device,
 			    struct drbd_peer_request *peer_req)
 {
+	bool use_sendpage = !(peer_req->flags & EE_RELEASE_TO_MEMPOOL);
 	struct page *page = peer_req->pages;
 	unsigned len = peer_req->i.size;
 	int err;
@@ -1619,8 +1610,13 @@ static int _drbd_send_zc_ee(struct drbd_peer_device *peer_device,
 	page_chain_for_each(page) {
 		unsigned l = min_t(unsigned, len, PAGE_SIZE);
 
-		err = _drbd_send_page(peer_device, page, 0, l,
-				      page_chain_next(page) ? MSG_MORE : 0);
+		if (likely(use_sendpage))
+			err = _drbd_send_page(peer_device, page, 0, l,
+					      page_chain_next(page) ? MSG_MORE : 0);
+		else
+			err = _drbd_no_send_page(peer_device, page, 0, l,
+						 page_chain_next(page) ? MSG_MORE : 0);
+
 		if (err)
 			return err;
 		len -= l;
@@ -1962,7 +1958,6 @@ void drbd_init_set_defaults(struct drbd_device *device)
 	INIT_LIST_HEAD(&device->sync_ee);
 	INIT_LIST_HEAD(&device->done_ee);
 	INIT_LIST_HEAD(&device->read_ee);
-	INIT_LIST_HEAD(&device->net_ee);
 	INIT_LIST_HEAD(&device->resync_reads);
 	INIT_LIST_HEAD(&device->resync_work.list);
 	INIT_LIST_HEAD(&device->unplug_work.list);
@@ -2043,7 +2038,6 @@ void drbd_device_cleanup(struct drbd_device *device)
 	D_ASSERT(device, list_empty(&device->sync_ee));
 	D_ASSERT(device, list_empty(&device->done_ee));
 	D_ASSERT(device, list_empty(&device->read_ee));
-	D_ASSERT(device, list_empty(&device->net_ee));
 	D_ASSERT(device, list_empty(&device->resync_reads));
 	D_ASSERT(device, list_empty(&first_peer_device(device)->connection->sender_work.q));
 	D_ASSERT(device, list_empty(&device->resync_work.list));
@@ -2055,19 +2049,11 @@ void drbd_device_cleanup(struct drbd_device *device)
 
 static void drbd_destroy_mempools(void)
 {
-	struct page *page;
-
-	while (drbd_pp_pool) {
-		page = drbd_pp_pool;
-		drbd_pp_pool = (struct page *)page_private(page);
-		__free_page(page);
-		drbd_pp_vacant--;
-	}
-
 	/* D_ASSERT(device, atomic_read(&drbd_pp_vacant)==0); */
 
 	bioset_exit(&drbd_io_bio_set);
 	bioset_exit(&drbd_md_io_bio_set);
+	mempool_exit(&drbd_buffer_page_pool);
 	mempool_exit(&drbd_md_io_page_pool);
 	mempool_exit(&drbd_ee_mempool);
 	mempool_exit(&drbd_request_mempool);
@@ -2086,9 +2072,8 @@ static void drbd_destroy_mempools(void)
 
 static int drbd_create_mempools(void)
 {
-	struct page *page;
 	const int number = (DRBD_MAX_BIO_SIZE/PAGE_SIZE) * drbd_minor_count;
-	int i, ret;
+	int ret;
 
 	/* caches */
 	drbd_request_cache = kmem_cache_create(
@@ -2125,6 +2110,10 @@ static int drbd_create_mempools(void)
 	if (ret)
 		goto Enomem;
 
+	ret = mempool_init_page_pool(&drbd_buffer_page_pool, number, 0);
+	if (ret)
+		goto Enomem;
+
 	ret = mempool_init_slab_pool(&drbd_request_mempool, number,
 				     drbd_request_cache);
 	if (ret)
@@ -2134,15 +2123,6 @@ static int drbd_create_mempools(void)
 	if (ret)
 		goto Enomem;
 
-	for (i = 0; i < number; i++) {
-		page = alloc_page(GFP_HIGHUSER);
-		if (!page)
-			goto Enomem;
-		set_page_private(page, (unsigned long)drbd_pp_pool);
-		drbd_pp_pool = page;
-	}
-	drbd_pp_vacant = number;
-
 	return 0;
 
 Enomem:
@@ -2169,10 +2149,6 @@ static void drbd_release_all_peer_reqs(struct drbd_device *device)
 	rr = drbd_free_peer_reqs(device, &device->done_ee);
 	if (rr)
 		drbd_err(device, "%d EEs in done list found!\n", rr);
-
-	rr = drbd_free_peer_reqs(device, &device->net_ee);
-	if (rr)
-		drbd_err(device, "%d EEs in net list found!\n", rr);
 }
 
 /* caution. no locking. */
@@ -2863,11 +2839,6 @@ static int __init drbd_init(void)
 		return err;
 	}
 
-	/*
-	 * allocate all necessary structs
-	 */
-	init_waitqueue_head(&drbd_pp_wait);
-
 	drbd_proc = NULL; /* play safe for drbd_cleanup */
 	idr_init(&drbd_devices);