Merge branch 'for-6.11/block-limits' into for-6.11/block

Merge in last round of queue limits changes from Christoph.

* for-6.11/block-limits: (26 commits)
  block: move the bounce flag into the features field
  block: move the skip_tagset_quiesce flag to queue_limits
  block: move the pci_p2pdma flag to queue_limits
  block: move the zone_resetall flag to queue_limits
  block: move the zoned flag into the features field
  block: move the poll flag to queue_limits
  block: move the dax flag to queue_limits
  block: move the nowait flag to queue_limits
  block: move the synchronous flag to queue_limits
  block: move the stable_writes flag to queue_limits
  block: move the io_stat flag setting to queue_limits
  block: move the add_random flag to queue_limits
  block: move the nonrot flag to queue_limits
  block: move cache control settings out of queue->flags
  block: remove blk_flush_policy
  block: freeze the queue in queue_attr_store
  nbd: move setting the cache control flags to __nbd_set_size
  virtio_blk: remove virtblk_update_cache_mode
  loop: fold loop_update_rotational into loop_reconfigure_limits
  loop: also use the default block size from an underlying block device
  ...

Signed-off-by: Jens Axboe <axboe@kernel.dk>

Author: axboe

Documentation/block/writeback_cache_control.rst

@@ -46,41 +46,50 @@ worry if the underlying devices need any explicit cache flushing and how
 the Forced Unit Access is implemented.  The REQ_PREFLUSH and REQ_FUA flags
 may both be set on a single bio.
 
+Feature settings for block drivers
+----------------------------------
 
-Implementation details for bio based block drivers
---------------------------------------------------------------
+For devices that do not support volatile write caches there is no driver
+support required, the block layer completes empty REQ_PREFLUSH requests before
+entering the driver and strips off the REQ_PREFLUSH and REQ_FUA bits from
+requests that have a payload.
 
-These drivers will always see the REQ_PREFLUSH and REQ_FUA bits as they sit
-directly below the submit_bio interface.  For remapping drivers the REQ_FUA
-bits need to be propagated to underlying devices, and a global flush needs
-to be implemented for bios with the REQ_PREFLUSH bit set.  For real device
-drivers that do not have a volatile cache the REQ_PREFLUSH and REQ_FUA bits
-on non-empty bios can simply be ignored, and REQ_PREFLUSH requests without
-data can be completed successfully without doing any work.  Drivers for
-devices with volatile caches need to implement the support for these
-flags themselves without any help from the block layer.
+For devices with volatile write caches the driver needs to tell the block layer
+that it supports flushing caches by setting the
 
+   BLK_FEAT_WRITE_CACHE
 
-Implementation details for request_fn based block drivers
----------------------------------------------------------
+flag in the queue_limits feature field.  For devices that also support the FUA
+bit the block layer needs to be told to pass on the REQ_FUA bit by also setting
+the
 
-For devices that do not support volatile write caches there is no driver
-support required, the block layer completes empty REQ_PREFLUSH requests before
-entering the driver and strips off the REQ_PREFLUSH and REQ_FUA bits from
-requests that have a payload.  For devices with volatile write caches the
-driver needs to tell the block layer that it supports flushing caches by
-doing::
+   BLK_FEAT_FUA
+
+flag in the features field of the queue_limits structure.
+
+Implementation details for bio based block drivers
+--------------------------------------------------
+
+For bio based drivers the REQ_PREFLUSH and REQ_FUA bit are simplify passed on
+to the driver if the drivers sets the BLK_FEAT_WRITE_CACHE flag and the drivers
+needs to handle them.
+
+*NOTE*: The REQ_FUA bit also gets passed on when the BLK_FEAT_FUA flags is
+_not_ set.  Any bio based driver that sets BLK_FEAT_WRITE_CACHE also needs to
+handle REQ_FUA.
 
-	blk_queue_write_cache(sdkp->disk->queue, true, false);
+For remapping drivers the REQ_FUA bits need to be propagated to underlying
+devices, and a global flush needs to be implemented for bios with the
+REQ_PREFLUSH bit set.
 
-and handle empty REQ_OP_FLUSH requests in its prep_fn/request_fn.  Note that
-REQ_PREFLUSH requests with a payload are automatically turned into a sequence
-of an empty REQ_OP_FLUSH request followed by the actual write by the block
-layer.  For devices that also support the FUA bit the block layer needs
-to be told to pass through the REQ_FUA bit using::
+Implementation details for blk-mq drivers
+-----------------------------------------
 
-	blk_queue_write_cache(sdkp->disk->queue, true, true);
+When the BLK_FEAT_WRITE_CACHE flag is set, REQ_OP_WRITE | REQ_PREFLUSH requests
+with a payload are automatically turned into a sequence of a REQ_OP_FLUSH
+request followed by the actual write by the block layer.
 
-and the driver must handle write requests that have the REQ_FUA bit set
-in prep_fn/request_fn.  If the FUA bit is not natively supported the block
-layer turns it into an empty REQ_OP_FLUSH request after the actual write.
+When the BLK_FEAT_FUA flags is set, the REQ_FUA bit simplify passed on for the
+REQ_OP_WRITE request, else a REQ_OP_FLUSH request is sent by the block layer
+after the completion of the write request for bio submissions with the REQ_FUA
+bit set.

arch/m68k/emu/nfblock.c

@@ -98,6 +98,7 @@ static int __init nfhd_init_one(int id, u32 blocks, u32 bsize)
 {
 	struct queue_limits lim = {
 		.logical_block_size	= bsize,
+		.features		= BLK_FEAT_ROTATIONAL,
 	};
 	struct nfhd_device *dev;
 	int dev_id = id - NFHD_DEV_OFFSET;

arch/um/drivers/ubd_kern.c

@@ -835,6 +835,7 @@ static int ubd_add(int n, char **error_out)
 	struct queue_limits lim = {
 		.max_segments		= MAX_SG,
 		.seg_boundary_mask	= PAGE_SIZE - 1,
+		.features		= BLK_FEAT_WRITE_CACHE,
 	};
 	struct gendisk *disk;
 	int err = 0;
@@ -881,8 +882,6 @@ static int ubd_add(int n, char **error_out)
 		goto out_cleanup_tags;
 	}
 
-	blk_queue_flag_set(QUEUE_FLAG_NONROT, disk->queue);
-	blk_queue_write_cache(disk->queue, true, false);
 	disk->major = UBD_MAJOR;
 	disk->first_minor = n << UBD_SHIFT;
 	disk->minors = 1 << UBD_SHIFT;

arch/xtensa/platforms/iss/simdisk.c

@@ -263,6 +263,9 @@ static const struct proc_ops simdisk_proc_ops = {
 static int __init simdisk_setup(struct simdisk *dev, int which,
 		struct proc_dir_entry *procdir)
 {
+	struct queue_limits lim = {
+		.features		= BLK_FEAT_ROTATIONAL,
+	};
 	char tmp[2] = { '0' + which, 0 };
 	int err;
 
@@ -271,7 +274,7 @@ static int __init simdisk_setup(struct simdisk *dev, int which,
 	spin_lock_init(&dev->lock);
 	dev->users = 0;
 
-	dev->gd = blk_alloc_disk(NULL, NUMA_NO_NODE);
+	dev->gd = blk_alloc_disk(&lim, NUMA_NO_NODE);
 	if (IS_ERR(dev->gd)) {
 		err = PTR_ERR(dev->gd);
 		goto out;

block/blk-core.c

@@ -782,7 +782,7 @@ void submit_bio_noacct(struct bio *bio)
 		if (WARN_ON_ONCE(bio_op(bio) != REQ_OP_WRITE &&
 				 bio_op(bio) != REQ_OP_ZONE_APPEND))
 			goto end_io;
-		if (!test_bit(QUEUE_FLAG_WC, &q->queue_flags)) {
+		if (!bdev_write_cache(bdev)) {
 			bio->bi_opf &= ~(REQ_PREFLUSH | REQ_FUA);
 			if (!bio_sectors(bio)) {
 				status = BLK_STS_OK;
@@ -791,7 +791,7 @@ void submit_bio_noacct(struct bio *bio)
 		}
 	}
 
-	if (!test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
+	if (!(q->limits.features & BLK_FEAT_POLL))
 		bio_clear_polled(bio);
 
 	switch (bio_op(bio)) {
@@ -915,8 +915,7 @@ int bio_poll(struct bio *bio, struct io_comp_batch *iob, unsigned int flags)
 		return 0;
 
 	q = bdev_get_queue(bdev);
-	if (cookie == BLK_QC_T_NONE ||
-	    !test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
+	if (cookie == BLK_QC_T_NONE || !(q->limits.features & BLK_FEAT_POLL))
 		return 0;
 
 	blk_flush_plug(current->plug, false);

block/blk-flush.c

@@ -100,23 +100,6 @@ blk_get_flush_queue(struct request_queue *q, struct blk_mq_ctx *ctx)
 	return blk_mq_map_queue(q, REQ_OP_FLUSH, ctx)->fq;
 }
 
-static unsigned int blk_flush_policy(unsigned long fflags, struct request *rq)
-{
-	unsigned int policy = 0;
-
-	if (blk_rq_sectors(rq))
-		policy |= REQ_FSEQ_DATA;
-
-	if (fflags & (1UL << QUEUE_FLAG_WC)) {
-		if (rq->cmd_flags & REQ_PREFLUSH)
-			policy |= REQ_FSEQ_PREFLUSH;
-		if (!(fflags & (1UL << QUEUE_FLAG_FUA)) &&
-		    (rq->cmd_flags & REQ_FUA))
-			policy |= REQ_FSEQ_POSTFLUSH;
-	}
-	return policy;
-}
-
 static unsigned int blk_flush_cur_seq(struct request *rq)
 {
 	return 1 << ffz(rq->flush.seq);
@@ -398,19 +381,32 @@ static void blk_rq_init_flush(struct request *rq)
 bool blk_insert_flush(struct request *rq)
 {
 	struct request_queue *q = rq->q;
-	unsigned long fflags = q->queue_flags;	/* may change, cache */
-	unsigned int policy = blk_flush_policy(fflags, rq);
 	struct blk_flush_queue *fq = blk_get_flush_queue(q, rq->mq_ctx);
+	bool supports_fua = q->limits.features & BLK_FEAT_FUA;
+	unsigned int policy = 0;
 
 	/* FLUSH/FUA request must never be merged */
 	WARN_ON_ONCE(rq->bio != rq->biotail);
 
+	if (blk_rq_sectors(rq))
+		policy |= REQ_FSEQ_DATA;
+
+	/*
+	 * Check which flushes we need to sequence for this operation.
+	 */
+	if (blk_queue_write_cache(q)) {
+		if (rq->cmd_flags & REQ_PREFLUSH)
+			policy |= REQ_FSEQ_PREFLUSH;
+		if ((rq->cmd_flags & REQ_FUA) && !supports_fua)
+			policy |= REQ_FSEQ_POSTFLUSH;
+	}
+
 	/*
 	 * @policy now records what operations need to be done.  Adjust
 	 * REQ_PREFLUSH and FUA for the driver.
 	 */
 	rq->cmd_flags &= ~REQ_PREFLUSH;
-	if (!(fflags & (1UL << QUEUE_FLAG_FUA)))
+	if (!supports_fua)
 		rq->cmd_flags &= ~REQ_FUA;
 
 	/*

block/blk-mq-debugfs.c

@@ -84,28 +84,15 @@ static const char *const blk_queue_flag_name[] = {
 	QUEUE_FLAG_NAME(NOMERGES),
 	QUEUE_FLAG_NAME(SAME_COMP),
 	QUEUE_FLAG_NAME(FAIL_IO),
-	QUEUE_FLAG_NAME(NONROT),
-	QUEUE_FLAG_NAME(IO_STAT),
 	QUEUE_FLAG_NAME(NOXMERGES),
-	QUEUE_FLAG_NAME(ADD_RANDOM),
-	QUEUE_FLAG_NAME(SYNCHRONOUS),
 	QUEUE_FLAG_NAME(SAME_FORCE),
 	QUEUE_FLAG_NAME(INIT_DONE),
-	QUEUE_FLAG_NAME(STABLE_WRITES),
-	QUEUE_FLAG_NAME(POLL),
-	QUEUE_FLAG_NAME(WC),
-	QUEUE_FLAG_NAME(FUA),
-	QUEUE_FLAG_NAME(DAX),
 	QUEUE_FLAG_NAME(STATS),
 	QUEUE_FLAG_NAME(REGISTERED),
 	QUEUE_FLAG_NAME(QUIESCED),
-	QUEUE_FLAG_NAME(PCI_P2PDMA),
-	QUEUE_FLAG_NAME(ZONE_RESETALL),
 	QUEUE_FLAG_NAME(RQ_ALLOC_TIME),
 	QUEUE_FLAG_NAME(HCTX_ACTIVE),
-	QUEUE_FLAG_NAME(NOWAIT),
 	QUEUE_FLAG_NAME(SQ_SCHED),
-	QUEUE_FLAG_NAME(SKIP_TAGSET_QUIESCE),
 };
 #undef QUEUE_FLAG_NAME
 

block/blk-mq.c

@@ -4109,14 +4109,26 @@ void blk_mq_release(struct request_queue *q)
 	blk_mq_sysfs_deinit(q);
 }
 
+static bool blk_mq_can_poll(struct blk_mq_tag_set *set)
+{
+	return set->nr_maps > HCTX_TYPE_POLL &&
+		set->map[HCTX_TYPE_POLL].nr_queues;
+}
+
 struct request_queue *blk_mq_alloc_queue(struct blk_mq_tag_set *set,
 		struct queue_limits *lim, void *queuedata)
 {
 	struct queue_limits default_lim = { };
 	struct request_queue *q;
 	int ret;
 
-	q = blk_alloc_queue(lim ? lim : &default_lim, set->numa_node);
+	if (!lim)
+		lim = &default_lim;
+	lim->features |= BLK_FEAT_IO_STAT | BLK_FEAT_NOWAIT;
+	if (blk_mq_can_poll(set))
+		lim->features |= BLK_FEAT_POLL;
+
+	q = blk_alloc_queue(lim, set->numa_node);
 	if (IS_ERR(q))
 		return q;
 	q->queuedata = queuedata;
@@ -4269,17 +4281,6 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 	mutex_unlock(&q->sysfs_lock);
 }
 
-static void blk_mq_update_poll_flag(struct request_queue *q)
-{
-	struct blk_mq_tag_set *set = q->tag_set;
-
-	if (set->nr_maps > HCTX_TYPE_POLL &&
-	    set->map[HCTX_TYPE_POLL].nr_queues)
-		blk_queue_flag_set(QUEUE_FLAG_POLL, q);
-	else
-		blk_queue_flag_clear(QUEUE_FLAG_POLL, q);
-}
-
 int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 		struct request_queue *q)
 {
@@ -4307,7 +4308,6 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	q->tag_set = set;
 
 	q->queue_flags |= QUEUE_FLAG_MQ_DEFAULT;
-	blk_mq_update_poll_flag(q);
 
 	INIT_DELAYED_WORK(&q->requeue_work, blk_mq_requeue_work);
 	INIT_LIST_HEAD(&q->flush_list);
@@ -4631,13 +4631,15 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
 	int ret;
 	unsigned long i;
 
+	if (WARN_ON_ONCE(!q->mq_freeze_depth))
+		return -EINVAL;
+
 	if (!set)
 		return -EINVAL;
 
 	if (q->nr_requests == nr)
 		return 0;
 
-	blk_mq_freeze_queue(q);
 	blk_mq_quiesce_queue(q);
 
 	ret = 0;
@@ -4671,7 +4673,6 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr)
 	}
 
 	blk_mq_unquiesce_queue(q);
-	blk_mq_unfreeze_queue(q);
 
 	return ret;
 }
@@ -4793,8 +4794,10 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
 fallback:
 	blk_mq_update_queue_map(set);
 	list_for_each_entry(q, &set->tag_list, tag_set_list) {
+		struct queue_limits lim;
+
 		blk_mq_realloc_hw_ctxs(set, q);
-		blk_mq_update_poll_flag(q);
+
 		if (q->nr_hw_queues != set->nr_hw_queues) {
 			int i = prev_nr_hw_queues;
 
@@ -4806,6 +4809,13 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
 			set->nr_hw_queues = prev_nr_hw_queues;
 			goto fallback;
 		}
+		lim = queue_limits_start_update(q);
+		if (blk_mq_can_poll(set))
+			lim.features |= BLK_FEAT_POLL;
+		else
+			lim.features &= ~BLK_FEAT_POLL;
+		if (queue_limits_commit_update(q, &lim) < 0)
+			pr_warn("updating the poll flag failed\n");
 		blk_mq_map_swqueue(q);
 	}