[Ocfs2-devel] use block_device based APIs in block layer consumers v2

[Ocfs2-devel] use block_device based APIs in block layer consumers v2

[Christoph Hellwig via Ocfs2-devel]

Hi Jens,

this series cleanups up the block layer API so that APIs consumed
by file systems are (almost) only struct block_devic based, so that
file systems don't have to poke into block layer internals like the
request_queue.

I also found a bunch of existing bugs related to partition offsets
and discard so these are fixed while going along.


Changes since v1:
 - fix a bisection hazard
 - minor spelling fixes
 - reorder hunks between two patches to make the changes more obvious
 - reorder a patch to be earlier in the series to ease backporting


Diffstat:
 arch/um/drivers/ubd_kern.c           |    2 
 block/blk-core.c                     |    4 -
 block/blk-lib.c                      |  124 ++++++++++++++++++++---------------
 block/blk-mq-debugfs.c               |    2 
 block/blk-settings.c                 |   74 ++++++++++++++++++++
 block/blk.h                          |   14 ---
 block/fops.c                         |    2 
 block/genhd.c                        |    4 -
 block/ioctl.c                        |   48 ++++++++++---
 block/partitions/core.c              |   12 ---
 drivers/block/drbd/drbd_main.c       |   51 ++++++--------
 drivers/block/drbd/drbd_nl.c         |   94 +++++++++++---------------
 drivers/block/drbd/drbd_receiver.c   |   13 +--
 drivers/block/loop.c                 |   15 +---
 drivers/block/nbd.c                  |    5 -
 drivers/block/null_blk/main.c        |    1 
 drivers/block/rbd.c                  |    1 
 drivers/block/rnbd/rnbd-clt.c        |    6 -
 drivers/block/rnbd/rnbd-srv-dev.h    |    8 --
 drivers/block/rnbd/rnbd-srv.c        |    5 -
 drivers/block/virtio_blk.c           |    2 
 drivers/block/xen-blkback/blkback.c  |   15 ++--
 drivers/block/xen-blkback/xenbus.c   |    9 --
 drivers/block/xen-blkfront.c         |    7 -
 drivers/block/zram/zram_drv.c        |    1 
 drivers/md/bcache/alloc.c            |    2 
 drivers/md/bcache/request.c          |    4 -
 drivers/md/bcache/super.c            |    3 
 drivers/md/bcache/sysfs.c            |    2 
 drivers/md/dm-cache-target.c         |    9 --
 drivers/md/dm-clone-target.c         |    9 --
 drivers/md/dm-io.c                   |    2 
 drivers/md/dm-log-writes.c           |    3 
 drivers/md/dm-raid.c                 |    9 --
 drivers/md/dm-table.c                |   25 +------
 drivers/md/dm-thin.c                 |   15 ----
 drivers/md/dm.c                      |    3 
 drivers/md/md-linear.c               |   11 ---
 drivers/md/md.c                      |    5 -
 drivers/md/raid0.c                   |    7 -
 drivers/md/raid1.c                   |   18 -----
 drivers/md/raid10.c                  |   20 -----
 drivers/md/raid5-cache.c             |    8 +-
 drivers/md/raid5.c                   |   14 +--
 drivers/mmc/core/queue.c             |    3 
 drivers/mtd/mtd_blkdevs.c            |    1 
 drivers/nvme/host/core.c             |    6 -
 drivers/nvme/target/io-cmd-bdev.c    |    2 
 drivers/nvme/target/zns.c            |    3 
 drivers/s390/block/dasd_fba.c        |    1 
 drivers/scsi/sd.c                    |    2 
 drivers/target/target_core_device.c  |   20 ++---
 drivers/target/target_core_file.c    |   10 +-
 drivers/target/target_core_iblock.c  |   17 +---
 fs/btrfs/disk-io.c                   |    3 
 fs/btrfs/extent-tree.c               |    8 +-
 fs/btrfs/ioctl.c                     |   12 +--
 fs/btrfs/volumes.c                   |    4 -
 fs/btrfs/zoned.c                     |    3 
 fs/direct-io.c                       |   32 +--------
 fs/exfat/file.c                      |    5 -
 fs/exfat/super.c                     |   10 --
 fs/ext4/ioctl.c                      |   10 --
 fs/ext4/mballoc.c                    |   10 +-
 fs/ext4/super.c                      |   10 --
 fs/f2fs/f2fs.h                       |    3 
 fs/f2fs/file.c                       |   19 ++---
 fs/f2fs/segment.c                    |    8 --
 fs/fat/file.c                        |    5 -
 fs/fat/inode.c                       |   10 --
 fs/gfs2/rgrp.c                       |    7 -
 fs/iomap/direct-io.c                 |    3 
 fs/jbd2/journal.c                    |    9 --
 fs/jfs/ioctl.c                       |    5 -
 fs/jfs/super.c                       |    8 --
 fs/nilfs2/ioctl.c                    |    6 -
 fs/nilfs2/sufile.c                   |    4 -
 fs/nilfs2/the_nilfs.c                |    4 -
 fs/ntfs3/file.c                      |    6 -
 fs/ntfs3/super.c                     |   10 +-
 fs/ocfs2/ioctl.c                     |    5 -
 fs/super.c                           |    2 
 fs/xfs/xfs_discard.c                 |    8 +-
 fs/xfs/xfs_log_cil.c                 |    2 
 fs/xfs/xfs_super.c                   |   12 +--
 fs/zonefs/super.c                    |    3 
 include/linux/blkdev.h               |  112 +++++++++++--------------------
 include/target/target_core_backend.h |    4 -
 mm/swapfile.c                        |   31 ++------
 89 files changed, 493 insertions(+), 653 deletions(-)

_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 01/27] target: remove an incorrect unmap zeroes data deduction

[Christoph Hellwig via Ocfs2-devel]

For block devices, the SCSI target drivers implements UNMAP as calls to
blkdev_issue_discard, which does not guarantee zeroing just because
Write Zeroes is supported.

Note that this does not affect the file backed path which uses
fallocate to punch holes.

Fixes: 2237498f0b5c ("target/iblock: Convert WRITE_SAME to blkdev_issue_zeroout")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/target/target_core_device.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index 44bb380e7390c..fa866acef5bb2 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -850,7 +850,6 @@ bool target_configure_unmap_from_queue(struct se_dev_attrib *attrib,
 	attrib->unmap_granularity = q->limits.discard_granularity / block_size;
 	attrib->unmap_granularity_alignment = q->limits.discard_alignment /
 								block_size;
-	attrib->unmap_zeroes_data = !!(q->limits.max_write_zeroes_sectors);
 	return true;
 }
 EXPORT_SYMBOL(target_configure_unmap_from_queue);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 02/27] target: pass a block_device to target_configure_unmap_from_queue

[Christoph Hellwig via Ocfs2-devel]

The SCSI target drivers is a consumer of the block layer and shoul
d generally work on struct block_device.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/target/target_core_device.c  | 5 +++--
 drivers/target/target_core_file.c    | 7 ++++---
 drivers/target/target_core_iblock.c  | 2 +-
 include/target/target_core_backend.h | 4 ++--
 4 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index fa866acef5bb2..3a1ec705cd80b 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -834,9 +834,10 @@ struct se_device *target_alloc_device(struct se_hba *hba, const char *name)
  * in ATA and we need to set TPE=1
  */
 bool target_configure_unmap_from_queue(struct se_dev_attrib *attrib,
-				       struct request_queue *q)
+				       struct block_device *bdev)
 {
-	int block_size = queue_logical_block_size(q);
+	struct request_queue *q = bdev_get_queue(bdev);
+	int block_size = bdev_logical_block_size(bdev);
 
 	if (!blk_queue_discard(q))
 		return false;
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index 8190b840065f3..8d191fdc33217 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -134,10 +134,11 @@ static int fd_configure_device(struct se_device *dev)
 	 */
 	inode = file->f_mapping->host;
 	if (S_ISBLK(inode->i_mode)) {
-		struct request_queue *q = bdev_get_queue(I_BDEV(inode));
+		struct block_device *bdev = I_BDEV(inode);
+		struct request_queue *q = bdev_get_queue(bdev);
 		unsigned long long dev_size;
 
-		fd_dev->fd_block_size = bdev_logical_block_size(I_BDEV(inode));
+		fd_dev->fd_block_size = bdev_logical_block_size(bdev);
 		/*
 		 * Determine the number of bytes from i_size_read() minus
 		 * one (1) logical sector from underlying struct block_device
@@ -150,7 +151,7 @@ static int fd_configure_device(struct se_device *dev)
 			dev_size, div_u64(dev_size, fd_dev->fd_block_size),
 			fd_dev->fd_block_size);
 
-		if (target_configure_unmap_from_queue(&dev->dev_attrib, q))
+		if (target_configure_unmap_from_queue(&dev->dev_attrib, bdev))
 			pr_debug("IFILE: BLOCK Discard support available,"
 				 " disabled by default\n");
 		/*
diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c
index 87ede165ddba4..b886ce1770bfd 100644
--- a/drivers/target/target_core_iblock.c
+++ b/drivers/target/target_core_iblock.c
@@ -119,7 +119,7 @@ static int iblock_configure_device(struct se_device *dev)
 	dev->dev_attrib.hw_max_sectors = queue_max_hw_sectors(q);
 	dev->dev_attrib.hw_queue_depth = q->nr_requests;
 
-	if (target_configure_unmap_from_queue(&dev->dev_attrib, q))
+	if (target_configure_unmap_from_queue(&dev->dev_attrib, bd))
 		pr_debug("IBLOCK: BLOCK Discard support available,"
 			 " disabled by default\n");
 
diff --git a/include/target/target_core_backend.h b/include/target/target_core_backend.h
index 675f3a1fe6139..773963a1e0b53 100644
--- a/include/target/target_core_backend.h
+++ b/include/target/target_core_backend.h
@@ -14,7 +14,7 @@
 #define TRANSPORT_FLAG_PASSTHROUGH_ALUA		0x2
 #define TRANSPORT_FLAG_PASSTHROUGH_PGR          0x4
 
-struct request_queue;
+struct block_device;
 struct scatterlist;
 
 struct target_backend_ops {
@@ -117,7 +117,7 @@ sense_reason_t passthrough_parse_cdb(struct se_cmd *cmd,
 bool target_sense_desc_format(struct se_device *dev);
 sector_t target_to_linux_sector(struct se_device *dev, sector_t lb);
 bool target_configure_unmap_from_queue(struct se_dev_attrib *attrib,
-				       struct request_queue *q);
+				       struct block_device *bdev);
 
 static inline bool target_dev_configured(struct se_device *se_dev)
 {
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 03/27] target: fix discard alignment on partitions

[Christoph Hellwig via Ocfs2-devel]

Use the proper bdev_discard_alignment helper that accounts for partition
offsets.

Fixes: c66ac9db8d4a ("[SCSI] target: Add LIO target core v4.0.0-rc6")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/target/target_core_device.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index 3a1ec705cd80b..16e775bcf4a7c 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -849,8 +849,8 @@ bool target_configure_unmap_from_queue(struct se_dev_attrib *attrib,
 	 */
 	attrib->max_unmap_block_desc_count = 1;
 	attrib->unmap_granularity = q->limits.discard_granularity / block_size;
-	attrib->unmap_granularity_alignment = q->limits.discard_alignment /
-								block_size;
+	attrib->unmap_granularity_alignment =
+		bdev_discard_alignment(bdev) / block_size;
 	return true;
 }
 EXPORT_SYMBOL(target_configure_unmap_from_queue);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 04/27] drbd: remove assign_p_sizes_qlim

[Christoph Hellwig via Ocfs2-devel]

Fold each branch into its only caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/block/drbd/drbd_main.c | 47 +++++++++++++++-------------------
 1 file changed, 20 insertions(+), 27 deletions(-)

diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 9676a1d214bc5..1262fe1c33618 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -903,31 +903,6 @@ void drbd_gen_and_send_sync_uuid(struct drbd_peer_device *peer_device)
 	}
 }
 
-/* communicated if (agreed_features & DRBD_FF_WSAME) */
-static void
-assign_p_sizes_qlim(struct drbd_device *device, struct p_sizes *p,
-					struct request_queue *q)
-{
-	if (q) {
-		p->qlim->physical_block_size = cpu_to_be32(queue_physical_block_size(q));
-		p->qlim->logical_block_size = cpu_to_be32(queue_logical_block_size(q));
-		p->qlim->alignment_offset = cpu_to_be32(queue_alignment_offset(q));
-		p->qlim->io_min = cpu_to_be32(queue_io_min(q));
-		p->qlim->io_opt = cpu_to_be32(queue_io_opt(q));
-		p->qlim->discard_enabled = blk_queue_discard(q);
-		p->qlim->write_same_capable = 0;
-	} else {
-		q = device->rq_queue;
-		p->qlim->physical_block_size = cpu_to_be32(queue_physical_block_size(q));
-		p->qlim->logical_block_size = cpu_to_be32(queue_logical_block_size(q));
-		p->qlim->alignment_offset = 0;
-		p->qlim->io_min = cpu_to_be32(queue_io_min(q));
-		p->qlim->io_opt = cpu_to_be32(queue_io_opt(q));
-		p->qlim->discard_enabled = 0;
-		p->qlim->write_same_capable = 0;
-	}
-}
-
 int drbd_send_sizes(struct drbd_peer_device *peer_device, int trigger_reply, enum dds_flags flags)
 {
 	struct drbd_device *device = peer_device->device;
@@ -957,14 +932,32 @@ int drbd_send_sizes(struct drbd_peer_device *peer_device, int trigger_reply, enu
 		q_order_type = drbd_queue_order_type(device);
 		max_bio_size = queue_max_hw_sectors(q) << 9;
 		max_bio_size = min(max_bio_size, DRBD_MAX_BIO_SIZE);
-		assign_p_sizes_qlim(device, p, q);
+		p->qlim->physical_block_size =
+			cpu_to_be32(queue_physical_block_size(q));
+		p->qlim->logical_block_size =
+			cpu_to_be32(queue_logical_block_size(q));
+		p->qlim->alignment_offset =
+			cpu_to_be32(queue_alignment_offset(q));
+		p->qlim->io_min = cpu_to_be32(queue_io_min(q));
+		p->qlim->io_opt = cpu_to_be32(queue_io_opt(q));
+		p->qlim->discard_enabled = blk_queue_discard(q);
 		put_ldev(device);
 	} else {
+		struct request_queue *q = device->rq_queue;
+
+		p->qlim->physical_block_size =
+			cpu_to_be32(queue_physical_block_size(q));
+		p->qlim->logical_block_size =
+			cpu_to_be32(queue_logical_block_size(q));
+		p->qlim->alignment_offset = 0;
+		p->qlim->io_min = cpu_to_be32(queue_io_min(q));
+		p->qlim->io_opt = cpu_to_be32(queue_io_opt(q));
+		p->qlim->discard_enabled = 0;
+
 		d_size = 0;
 		u_size = 0;
 		q_order_type = QUEUE_ORDERED_NONE;
 		max_bio_size = DRBD_MAX_BIO_SIZE; /* ... multiple BIOs per peer_request */
-		assign_p_sizes_qlim(device, p, NULL);
 	}
 
 	if (peer_device->connection->agreed_pro_version <= 94)
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 05/27] drbd: use bdev based limit helpers in drbd_send_sizes

[Christoph Hellwig via Ocfs2-devel]

Use the bdev based limits helpers where they exist.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christoph Böhmwalder <christoph.boehmwalder@linbit.com>
---
 drivers/block/drbd/drbd_main.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 1262fe1c33618..aa2c21aeb747c 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -924,7 +924,9 @@ int drbd_send_sizes(struct drbd_peer_device *peer_device, int trigger_reply, enu
 
 	memset(p, 0, packet_size);
 	if (get_ldev_if_state(device, D_NEGOTIATING)) {
-		struct request_queue *q = bdev_get_queue(device->ldev->backing_bdev);
+		struct block_device *bdev = device->ldev->backing_bdev;
+		struct request_queue *q = bdev_get_queue(bdev);
+
 		d_size = drbd_get_max_capacity(device->ldev);
 		rcu_read_lock();
 		u_size = rcu_dereference(device->ldev->disk_conf)->disk_size;
@@ -933,13 +935,13 @@ int drbd_send_sizes(struct drbd_peer_device *peer_device, int trigger_reply, enu
 		max_bio_size = queue_max_hw_sectors(q) << 9;
 		max_bio_size = min(max_bio_size, DRBD_MAX_BIO_SIZE);
 		p->qlim->physical_block_size =
-			cpu_to_be32(queue_physical_block_size(q));
+			cpu_to_be32(bdev_physical_block_size(bdev));
 		p->qlim->logical_block_size =
-			cpu_to_be32(queue_logical_block_size(q));
+			cpu_to_be32(bdev_logical_block_size(bdev));
 		p->qlim->alignment_offset =
 			cpu_to_be32(queue_alignment_offset(q));
-		p->qlim->io_min = cpu_to_be32(queue_io_min(q));
-		p->qlim->io_opt = cpu_to_be32(queue_io_opt(q));
+		p->qlim->io_min = cpu_to_be32(bdev_io_min(bdev));
+		p->qlim->io_opt = cpu_to_be32(bdev_io_opt(bdev));
 		p->qlim->discard_enabled = blk_queue_discard(q);
 		put_ldev(device);
 	} else {
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 06/27] drbd: use bdev_alignment_offset instead of queue_alignment_offset

[Christoph Hellwig via Ocfs2-devel]

The bdev version does the right thing for partitions, so use that.

Fixes: 9104d31a759f ("drbd: introduce WRITE_SAME support")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christoph Böhmwalder <christoph.boehmwalder@linbit.com>
---
 drivers/block/drbd/drbd_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index aa2c21aeb747c..eae629c4f6eaf 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -939,7 +939,7 @@ int drbd_send_sizes(struct drbd_peer_device *peer_device, int trigger_reply, enu
 		p->qlim->logical_block_size =
 			cpu_to_be32(bdev_logical_block_size(bdev));
 		p->qlim->alignment_offset =
-			cpu_to_be32(queue_alignment_offset(q));
+			cpu_to_be32(bdev_alignment_offset(bdev));
 		p->qlim->io_min = cpu_to_be32(bdev_io_min(bdev));
 		p->qlim->io_opt = cpu_to_be32(bdev_io_opt(bdev));
 		p->qlim->discard_enabled = blk_queue_discard(q);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 07/27] drbd: cleanup decide_on_discard_support

[Christoph Hellwig via Ocfs2-devel]

Sanitize the calling conventions and use a goto label to cleanup the
code flow.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christoph Böhmwalder <christoph.boehmwalder@linbit.com>
---
 drivers/block/drbd/drbd_nl.c | 68 +++++++++++++++++++-----------------
 1 file changed, 35 insertions(+), 33 deletions(-)

diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 02030c9c4d3b1..40bb0b356a6d6 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -1204,38 +1204,42 @@ static unsigned int drbd_max_discard_sectors(struct drbd_connection *connection)
 }
 
 static void decide_on_discard_support(struct drbd_device *device,
-			struct request_queue *q,
-			struct request_queue *b,
-			bool discard_zeroes_if_aligned)
+		struct drbd_backing_dev *bdev)
 {
-	/* q = drbd device queue (device->rq_queue)
-	 * b = backing device queue (device->ldev->backing_bdev->bd_disk->queue),
-	 *     or NULL if diskless
-	 */
-	struct drbd_connection *connection = first_peer_device(device)->connection;
-	bool can_do = b ? blk_queue_discard(b) : true;
-
-	if (can_do && connection->cstate >= C_CONNECTED && !(connection->agreed_features & DRBD_FF_TRIM)) {
-		can_do = false;
-		drbd_info(connection, "peer DRBD too old, does not support TRIM: disabling discards\n");
-	}
-	if (can_do) {
-		/* We don't care for the granularity, really.
-		 * Stacking limits below should fix it for the local
-		 * device.  Whether or not it is a suitable granularity
-		 * on the remote device is not our problem, really. If
-		 * you care, you need to use devices with similar
-		 * topology on all peers. */
-		blk_queue_discard_granularity(q, 512);
-		q->limits.max_discard_sectors = drbd_max_discard_sectors(connection);
-		blk_queue_flag_set(QUEUE_FLAG_DISCARD, q);
-		q->limits.max_write_zeroes_sectors = drbd_max_discard_sectors(connection);
-	} else {
-		blk_queue_flag_clear(QUEUE_FLAG_DISCARD, q);
-		blk_queue_discard_granularity(q, 0);
-		q->limits.max_discard_sectors = 0;
-		q->limits.max_write_zeroes_sectors = 0;
+	struct drbd_connection *connection =
+		first_peer_device(device)->connection;
+	struct request_queue *q = device->rq_queue;
+
+	if (bdev && !blk_queue_discard(bdev->backing_bdev->bd_disk->queue))
+		goto not_supported;
+
+	if (connection->cstate >= C_CONNECTED &&
+	    !(connection->agreed_features & DRBD_FF_TRIM)) {
+		drbd_info(connection,
+			"peer DRBD too old, does not support TRIM: disabling discards\n");
+		goto not_supported;
 	}
+
+	/*
+	 * We don't care for the granularity, really.
+	 *
+	 * Stacking limits below should fix it for the local device.  Whether or
+	 * not it is a suitable granularity on the remote device is not our
+	 * problem, really. If you care, you need to use devices with similar
+	 * topology on all peers.
+	 */
+	blk_queue_discard_granularity(q, 512);
+	q->limits.max_discard_sectors = drbd_max_discard_sectors(connection);
+	blk_queue_flag_set(QUEUE_FLAG_DISCARD, q);
+	q->limits.max_write_zeroes_sectors =
+		drbd_max_discard_sectors(connection);
+	return;
+
+not_supported:
+	blk_queue_flag_clear(QUEUE_FLAG_DISCARD, q);
+	blk_queue_discard_granularity(q, 0);
+	q->limits.max_discard_sectors = 0;
+	q->limits.max_write_zeroes_sectors = 0;
 }
 
 static void fixup_discard_if_not_supported(struct request_queue *q)
@@ -1273,7 +1277,6 @@ static void drbd_setup_queue_param(struct drbd_device *device, struct drbd_backi
 	unsigned int max_segments = 0;
 	struct request_queue *b = NULL;
 	struct disk_conf *dc;
-	bool discard_zeroes_if_aligned = true;
 
 	if (bdev) {
 		b = bdev->backing_bdev->bd_disk->queue;
@@ -1282,7 +1285,6 @@ static void drbd_setup_queue_param(struct drbd_device *device, struct drbd_backi
 		rcu_read_lock();
 		dc = rcu_dereference(device->ldev->disk_conf);
 		max_segments = dc->max_bio_bvecs;
-		discard_zeroes_if_aligned = dc->discard_zeroes_if_aligned;
 		rcu_read_unlock();
 
 		blk_set_stacking_limits(&q->limits);
@@ -1292,7 +1294,7 @@ static void drbd_setup_queue_param(struct drbd_device *device, struct drbd_backi
 	/* This is the workaround for "bio would need to, but cannot, be split" */
 	blk_queue_max_segments(q, max_segments ? max_segments : BLK_MAX_SEGMENTS);
 	blk_queue_segment_boundary(q, PAGE_SIZE-1);
-	decide_on_discard_support(device, q, b, discard_zeroes_if_aligned);
+	decide_on_discard_support(device, bdev);
 
 	if (b) {
 		blk_stack_limits(&q->limits, &b->limits, 0);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 08/27] btrfs: use bdev_max_active_zones instead of open coding it

[Christoph Hellwig via Ocfs2-devel]

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
---
 fs/btrfs/zoned.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index 1b1b310c3c510..f72cad7391a11 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -350,7 +350,6 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
 	struct btrfs_fs_info *fs_info = device->fs_info;
 	struct btrfs_zoned_device_info *zone_info = NULL;
 	struct block_device *bdev = device->bdev;
-	struct request_queue *queue = bdev_get_queue(bdev);
 	unsigned int max_active_zones;
 	unsigned int nactive;
 	sector_t nr_sectors;
@@ -410,7 +409,7 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
 	if (!IS_ALIGNED(nr_sectors, zone_sectors))
 		zone_info->nr_zones++;
 
-	max_active_zones = queue_max_active_zones(queue);
+	max_active_zones = bdev_max_active_zones(bdev);
 	if (max_active_zones && max_active_zones < BTRFS_MIN_ACTIVE_ZONES) {
 		btrfs_err_in_rcu(fs_info,
 "zoned: %s: max active zones %u is too small, need at least %u active zones",
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 09/27] ntfs3: use bdev_logical_block_size instead of open coding it

[Christoph Hellwig via Ocfs2-devel]

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
---
 fs/ntfs3/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c
index 278dcf5024102..cd30e81abbce0 100644
--- a/fs/ntfs3/super.c
+++ b/fs/ntfs3/super.c
@@ -920,7 +920,7 @@ static int ntfs_fill_super(struct super_block *sb, struct fs_context *fc)
 	}
 
 	/* Parse boot. */
-	err = ntfs_init_from_boot(sb, rq ? queue_logical_block_size(rq) : 512,
+	err = ntfs_init_from_boot(sb, bdev_logical_block_size(bdev),
 				  bdev_nr_bytes(bdev));
 	if (err)
 		goto out;
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 10/27] mm: use bdev_is_zoned in claim_swapfile

[Christoph Hellwig via Ocfs2-devel]

Use the bdev based helper instead of poking into the queue.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 mm/swapfile.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index 63c61f8b26118..4c7537162af5e 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -2761,7 +2761,7 @@ static int claim_swapfile(struct swap_info_struct *p, struct inode *inode)
 		 * write only restriction.  Hence zoned block devices are not
 		 * suitable for swapping.  Disallow them here.
 		 */
-		if (blk_queue_is_zoned(p->bdev->bd_disk->queue))
+		if (bdev_is_zoned(p->bdev))
 			return -EINVAL;
 		p->flags |= SWP_BLKDEV;
 	} else if (S_ISREG(inode->i_mode)) {
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 13/27] block: add a bdev_fua helper

[Christoph Hellwig via Ocfs2-devel]

Add a helper to check the FUA flag based on the block_device instead of
having to poke into the block layer internal request_queue.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/block/rnbd/rnbd-srv.c       | 3 +--
 drivers/target/target_core_iblock.c | 3 +--
 fs/iomap/direct-io.c                | 3 +--
 include/linux/blkdev.h              | 6 +++++-
 4 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/block/rnbd/rnbd-srv.c b/drivers/block/rnbd/rnbd-srv.c
index f8cc3c5fecb4b..beaef43a67b9d 100644
--- a/drivers/block/rnbd/rnbd-srv.c
+++ b/drivers/block/rnbd/rnbd-srv.c
@@ -533,7 +533,6 @@ static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 					struct rnbd_srv_sess_dev *sess_dev)
 {
 	struct rnbd_dev *rnbd_dev = sess_dev->rnbd_dev;
-	struct request_queue *q = bdev_get_queue(rnbd_dev->bdev);
 
 	rsp->hdr.type = cpu_to_le16(RNBD_MSG_OPEN_RSP);
 	rsp->device_id =
@@ -560,7 +559,7 @@ static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 	rsp->cache_policy = 0;
 	if (bdev_write_cache(rnbd_dev->bdev))
 		rsp->cache_policy |= RNBD_WRITEBACK;
-	if (blk_queue_fua(q))
+	if (bdev_fua(rnbd_dev->bdev))
 		rsp->cache_policy |= RNBD_FUA;
 }
 
diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c
index 03013e85ffc03..c4a903b8a47fc 100644
--- a/drivers/target/target_core_iblock.c
+++ b/drivers/target/target_core_iblock.c
@@ -727,14 +727,13 @@ iblock_execute_rw(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents,
 
 	if (data_direction == DMA_TO_DEVICE) {
 		struct iblock_dev *ib_dev = IBLOCK_DEV(dev);
-		struct request_queue *q = bdev_get_queue(ib_dev->ibd_bd);
 		/*
 		 * Force writethrough using REQ_FUA if a volatile write cache
 		 * is not enabled, or if initiator set the Force Unit Access bit.
 		 */
 		opf = REQ_OP_WRITE;
 		miter_dir = SG_MITER_TO_SG;
-		if (test_bit(QUEUE_FLAG_FUA, &q->queue_flags)) {
+		if (bdev_fua(ib_dev->ibd_bd)) {
 			if (cmd->se_cmd_flags & SCF_FUA)
 				opf |= REQ_FUA;
 			else if (!bdev_write_cache(ib_dev->ibd_bd))
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index b08f5dc31780d..62da020d02a11 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -265,8 +265,7 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
 		 * cache flushes on IO completion.
 		 */
 		if (!(iomap->flags & (IOMAP_F_SHARED|IOMAP_F_DIRTY)) &&
-		    (dio->flags & IOMAP_DIO_WRITE_FUA) &&
-		    blk_queue_fua(bdev_get_queue(iomap->bdev)))
+		    (dio->flags & IOMAP_DIO_WRITE_FUA) && bdev_fua(iomap->bdev))
 			use_fua = true;
 	}
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 807a49aa5a27a..075b16d4560e7 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -602,7 +602,6 @@ bool blk_queue_flag_test_and_set(unsigned int flag, struct request_queue *q);
 			     REQ_FAILFAST_DRIVER))
 #define blk_queue_quiesced(q)	test_bit(QUEUE_FLAG_QUIESCED, &(q)->queue_flags)
 #define blk_queue_pm_only(q)	atomic_read(&(q)->pm_only)
-#define blk_queue_fua(q)	test_bit(QUEUE_FLAG_FUA, &(q)->queue_flags)
 #define blk_queue_registered(q)	test_bit(QUEUE_FLAG_REGISTERED, &(q)->queue_flags)
 #define blk_queue_nowait(q)	test_bit(QUEUE_FLAG_NOWAIT, &(q)->queue_flags)
 
@@ -1336,6 +1335,11 @@ static inline bool bdev_write_cache(struct block_device *bdev)
 	return test_bit(QUEUE_FLAG_WC, &bdev_get_queue(bdev)->queue_flags);
 }
 
+static inline bool bdev_fua(struct block_device *bdev)
+{
+	return test_bit(QUEUE_FLAG_FUA, &bdev_get_queue(bdev)->queue_flags);
+}
+
 static inline enum blk_zoned_model bdev_zoned_model(struct block_device *bdev)
 {
 	struct request_queue *q = bdev_get_queue(bdev);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 14/27] block: add a bdev_stable_writes helper

[Christoph Hellwig via Ocfs2-devel]

Add a helper to check the stable writes flag based on the block_device
instead of having to poke into the block layer internal request_queue.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/md/dm-table.c  | 4 +---
 fs/super.c             | 2 +-
 include/linux/blkdev.h | 6 ++++++
 mm/swapfile.c          | 2 +-
 4 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 5e38d0dd009d5..d46839faa0ca5 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1950,9 +1950,7 @@ static int device_requires_stable_pages(struct dm_target *ti,
 					struct dm_dev *dev, sector_t start,
 					sector_t len, void *data)
 {
-	struct request_queue *q = bdev_get_queue(dev->bdev);
-
-	return blk_queue_stable_writes(q);
+	return bdev_stable_writes(dev->bdev);
 }
 
 int dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
diff --git a/fs/super.c b/fs/super.c
index f1d4a193602d6..60f57c7bc0a69 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -1204,7 +1204,7 @@ static int set_bdev_super(struct super_block *s, void *data)
 	s->s_dev = s->s_bdev->bd_dev;
 	s->s_bdi = bdi_get(s->s_bdev->bd_disk->bdi);
 
-	if (blk_queue_stable_writes(s->s_bdev->bd_disk->queue))
+	if (bdev_stable_writes(s->s_bdev))
 		s->s_iflags |= SB_I_STABLE_WRITES;
 	return 0;
 }
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 075b16d4560e7..a433798c3343e 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1330,6 +1330,12 @@ static inline bool bdev_nonrot(struct block_device *bdev)
 	return blk_queue_nonrot(bdev_get_queue(bdev));
 }
 
+static inline bool bdev_stable_writes(struct block_device *bdev)
+{
+	return test_bit(QUEUE_FLAG_STABLE_WRITES,
+			&bdev_get_queue(bdev)->queue_flags);
+}
+
 static inline bool bdev_write_cache(struct block_device *bdev)
 {
 	return test_bit(QUEUE_FLAG_WC, &bdev_get_queue(bdev)->queue_flags);
diff --git a/mm/swapfile.c b/mm/swapfile.c
index d5ab7ec4d92ca..4069f17a82c8e 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -3065,7 +3065,7 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
 		goto bad_swap_unlock_inode;
 	}
 
-	if (p->bdev && blk_queue_stable_writes(p->bdev->bd_disk->queue))
+	if (p->bdev && bdev_stable_writes(p->bdev))
 		p->flags |= SWP_STABLE_WRITES;
 
 	if (p->bdev && p->bdev->bd_disk->fops->rw_page)
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 16/27] block: use bdev_alignment_offset in part_alignment_offset_show

[Christoph Hellwig via Ocfs2-devel]

Replace the open coded offset calculation with the proper helper.
This is an ABI change in that the -1 for a misaligned partition is
properly propagated, which can be considered a bug fix and matches
what is done on the whole device.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/partitions/core.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/block/partitions/core.c b/block/partitions/core.c
index 2ef8dfa1e5c85..240b3fff521e4 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -200,11 +200,7 @@ static ssize_t part_ro_show(struct device *dev,
 static ssize_t part_alignment_offset_show(struct device *dev,
 					  struct device_attribute *attr, char *buf)
 {
-	struct block_device *bdev = dev_to_bdev(dev);
-
-	return sprintf(buf, "%u\n",
-		queue_limit_alignment_offset(&bdev_get_queue(bdev)->limits,
-				bdev->bd_start_sect));
+	return sprintf(buf, "%u\n", bdev_alignment_offset(dev_to_bdev(dev)));
 }
 
 static ssize_t part_discard_alignment_show(struct device *dev,
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 17/27] block: use bdev_alignment_offset in disk_alignment_offset_show

[Christoph Hellwig via Ocfs2-devel]

This does the same as the open coded variant except for an extra branch,
and allows to remove queue_alignment_offset entirely.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/genhd.c          | 2 +-
 include/linux/blkdev.h | 8 --------
 2 files changed, 1 insertion(+), 9 deletions(-)

diff --git a/block/genhd.c b/block/genhd.c
index b8b6759d670f0..712031ce19070 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1010,7 +1010,7 @@ static ssize_t disk_alignment_offset_show(struct device *dev,
 {
 	struct gendisk *disk = dev_to_disk(dev);
 
-	return sprintf(buf, "%d\n", queue_alignment_offset(disk->queue));
+	return sprintf(buf, "%d\n", bdev_alignment_offset(disk->part0));
 }
 
 static ssize_t disk_discard_alignment_show(struct device *dev,
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index f8c50b77543eb..d5346e72e3645 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1251,14 +1251,6 @@ bdev_zone_write_granularity(struct block_device *bdev)
 	return queue_zone_write_granularity(bdev_get_queue(bdev));
 }
 
-static inline int queue_alignment_offset(const struct request_queue *q)
-{
-	if (q->limits.misaligned)
-		return -1;
-
-	return q->limits.alignment_offset;
-}
-
 static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
 {
 	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 18/27] block: move bdev_alignment_offset and queue_limit_alignment_offset out of line

[Christoph Hellwig via Ocfs2-devel]

No need to inline these fairly larger helpers.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/blk-settings.c   | 23 +++++++++++++++++++++++
 include/linux/blkdev.h | 21 +--------------------
 2 files changed, 24 insertions(+), 20 deletions(-)

diff --git a/block/blk-settings.c b/block/blk-settings.c
index b83df3d2eebca..94410a13c0dee 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -468,6 +468,16 @@ void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
 }
 EXPORT_SYMBOL(blk_queue_io_opt);
 
+static int queue_limit_alignment_offset(struct queue_limits *lim,
+		sector_t sector)
+{
+	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
+	unsigned int alignment = sector_div(sector, granularity >> SECTOR_SHIFT)
+		<< SECTOR_SHIFT;
+
+	return (granularity + lim->alignment_offset - alignment) % granularity;
+}
+
 static unsigned int blk_round_down_sectors(unsigned int sectors, unsigned int lbs)
 {
 	sectors = round_down(sectors, lbs >> SECTOR_SHIFT);
@@ -901,3 +911,16 @@ void blk_queue_set_zoned(struct gendisk *disk, enum blk_zoned_model model)
 	}
 }
 EXPORT_SYMBOL_GPL(blk_queue_set_zoned);
+
+int bdev_alignment_offset(struct block_device *bdev)
+{
+	struct request_queue *q = bdev_get_queue(bdev);
+
+	if (q->limits.misaligned)
+		return -1;
+	if (bdev_is_partition(bdev))
+		return queue_limit_alignment_offset(&q->limits,
+				bdev->bd_start_sect);
+	return q->limits.alignment_offset;
+}
+EXPORT_SYMBOL_GPL(bdev_alignment_offset);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d5346e72e3645..0a1795ac26275 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1251,26 +1251,7 @@ bdev_zone_write_granularity(struct block_device *bdev)
 	return queue_zone_write_granularity(bdev_get_queue(bdev));
 }
 
-static inline int queue_limit_alignment_offset(struct queue_limits *lim, sector_t sector)
-{
-	unsigned int granularity = max(lim->physical_block_size, lim->io_min);
-	unsigned int alignment = sector_div(sector, granularity >> SECTOR_SHIFT)
-		<< SECTOR_SHIFT;
-
-	return (granularity + lim->alignment_offset - alignment) % granularity;
-}
-
-static inline int bdev_alignment_offset(struct block_device *bdev)
-{
-	struct request_queue *q = bdev_get_queue(bdev);
-
-	if (q->limits.misaligned)
-		return -1;
-	if (bdev_is_partition(bdev))
-		return queue_limit_alignment_offset(&q->limits,
-				bdev->bd_start_sect);
-	return q->limits.alignment_offset;
-}
+int bdev_alignment_offset(struct block_device *bdev);
 
 static inline int queue_discard_alignment(const struct request_queue *q)
 {
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 19/27] block: remove queue_discard_alignment

[Christoph Hellwig via Ocfs2-devel]

Just use bdev_alignment_offset in disk_discard_alignment_show instead.
That helpers is the same except for an always false branch that doesn't
matter in this slow path.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/genhd.c          | 2 +-
 include/linux/blkdev.h | 8 --------
 2 files changed, 1 insertion(+), 9 deletions(-)

diff --git a/block/genhd.c b/block/genhd.c
index 712031ce19070..36532b9318419 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1019,7 +1019,7 @@ static ssize_t disk_discard_alignment_show(struct device *dev,
 {
 	struct gendisk *disk = dev_to_disk(dev);
 
-	return sprintf(buf, "%d\n", queue_discard_alignment(disk->queue));
+	return sprintf(buf, "%d\n", bdev_alignment_offset(disk->part0));
 }
 
 static ssize_t diskseq_show(struct device *dev,
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0a1795ac26275..5a9b7aeda010b 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1253,14 +1253,6 @@ bdev_zone_write_granularity(struct block_device *bdev)
 
 int bdev_alignment_offset(struct block_device *bdev);
 
-static inline int queue_discard_alignment(const struct request_queue *q)
-{
-	if (q->limits.discard_misaligned)
-		return -1;
-
-	return q->limits.discard_alignment;
-}
-
 static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
 {
 	unsigned int alignment, granularity, offset;
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 20/27] block: use bdev_discard_alignment in part_discard_alignment_show

[Christoph Hellwig via Ocfs2-devel]

Use the bdev based alignment helper instead of open coding it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/partitions/core.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/block/partitions/core.c b/block/partitions/core.c
index 240b3fff521e4..70dec1c78521d 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -206,11 +206,7 @@ static ssize_t part_alignment_offset_show(struct device *dev,
 static ssize_t part_discard_alignment_show(struct device *dev,
 					   struct device_attribute *attr, char *buf)
 {
-	struct block_device *bdev = dev_to_bdev(dev);
-
-	return sprintf(buf, "%u\n",
-		queue_limit_discard_alignment(&bdev_get_queue(bdev)->limits,
-				bdev->bd_start_sect));
+	return sprintf(buf, "%u\n", bdev_discard_alignment(dev_to_bdev(dev)));
 }
 
 static DEVICE_ATTR(partition, 0444, part_partition_show, NULL);
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 21/27] block: move {bdev, queue_limit}_discard_alignment out of line

[Christoph Hellwig via Ocfs2-devel]

No need to inline these fairly larger helpers.  Also fix the return value
to be unsigned, just like the field in struct queue_limits.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/blk-settings.c   | 35 +++++++++++++++++++++++++++++++++++
 include/linux/blkdev.h | 34 +---------------------------------
 2 files changed, 36 insertions(+), 33 deletions(-)

diff --git a/block/blk-settings.c b/block/blk-settings.c
index 94410a13c0dee..fd83d674afd0a 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -478,6 +478,30 @@ static int queue_limit_alignment_offset(struct queue_limits *lim,
 	return (granularity + lim->alignment_offset - alignment) % granularity;
 }
 
+static unsigned int queue_limit_discard_alignment(struct queue_limits *lim,
+		sector_t sector)
+{
+	unsigned int alignment, granularity, offset;
+
+	if (!lim->max_discard_sectors)
+		return 0;
+
+	/* Why are these in bytes, not sectors? */
+	alignment = lim->discard_alignment >> SECTOR_SHIFT;
+	granularity = lim->discard_granularity >> SECTOR_SHIFT;
+	if (!granularity)
+		return 0;
+
+	/* Offset of the partition start in 'granularity' sectors */
+	offset = sector_div(sector, granularity);
+
+	/* And why do we do this modulus *again* in blkdev_issue_discard()? */
+	offset = (granularity + alignment - offset) % granularity;
+
+	/* Turn it back into bytes, gaah */
+	return offset << SECTOR_SHIFT;
+}
+
 static unsigned int blk_round_down_sectors(unsigned int sectors, unsigned int lbs)
 {
 	sectors = round_down(sectors, lbs >> SECTOR_SHIFT);
@@ -924,3 +948,14 @@ int bdev_alignment_offset(struct block_device *bdev)
 	return q->limits.alignment_offset;
 }
 EXPORT_SYMBOL_GPL(bdev_alignment_offset);
+
+unsigned int bdev_discard_alignment(struct block_device *bdev)
+{
+	struct request_queue *q = bdev_get_queue(bdev);
+
+	if (bdev_is_partition(bdev))
+		return queue_limit_discard_alignment(&q->limits,
+				bdev->bd_start_sect);
+	return q->limits.discard_alignment;
+}
+EXPORT_SYMBOL_GPL(bdev_discard_alignment);
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 5a9b7aeda010b..34b1cfd067421 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1252,39 +1252,7 @@ bdev_zone_write_granularity(struct block_device *bdev)
 }
 
 int bdev_alignment_offset(struct block_device *bdev);
-
-static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector_t sector)
-{
-	unsigned int alignment, granularity, offset;
-
-	if (!lim->max_discard_sectors)
-		return 0;
-
-	/* Why are these in bytes, not sectors? */
-	alignment = lim->discard_alignment >> SECTOR_SHIFT;
-	granularity = lim->discard_granularity >> SECTOR_SHIFT;
-	if (!granularity)
-		return 0;
-
-	/* Offset of the partition start in 'granularity' sectors */
-	offset = sector_div(sector, granularity);
-
-	/* And why do we do this modulus *again* in blkdev_issue_discard()? */
-	offset = (granularity + alignment - offset) % granularity;
-
-	/* Turn it back into bytes, gaah */
-	return offset << SECTOR_SHIFT;
-}
-
-static inline int bdev_discard_alignment(struct block_device *bdev)
-{
-	struct request_queue *q = bdev_get_queue(bdev);
-
-	if (bdev_is_partition(bdev))
-		return queue_limit_discard_alignment(&q->limits,
-				bdev->bd_start_sect);
-	return q->limits.discard_alignment;
-}
+unsigned int bdev_discard_alignment(struct block_device *bdev);
 
 static inline unsigned int bdev_write_zeroes_sectors(struct block_device *bdev)
 {
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 27/27] direct-io: remove random prefetches

[Christoph Hellwig via Ocfs2-devel]

Randomly poking into block device internals for manual prefetches isn't
exactly a very maintainable thing to do.  And none of the performance
criticil direct I/O implementations still use this library function
anyway, so just drop it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/direct-io.c | 32 ++++----------------------------
 1 file changed, 4 insertions(+), 28 deletions(-)

diff --git a/fs/direct-io.c b/fs/direct-io.c
index aef06e607b405..840752006f601 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -1115,11 +1115,10 @@ static inline int drop_refcount(struct dio *dio)
  * individual fields and will generate much worse code. This is important
  * for the whole file.
  */
-static inline ssize_t
-do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
-		      struct block_device *bdev, struct iov_iter *iter,
-		      get_block_t get_block, dio_iodone_t end_io,
-		      dio_submit_t submit_io, int flags)
+ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
+		struct block_device *bdev, struct iov_iter *iter,
+		get_block_t get_block, dio_iodone_t end_io,
+		dio_submit_t submit_io, int flags)
 {
 	unsigned i_blkbits = READ_ONCE(inode->i_blkbits);
 	unsigned blkbits = i_blkbits;
@@ -1334,29 +1333,6 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
 	kmem_cache_free(dio_cache, dio);
 	return retval;
 }
-
-ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
-			     struct block_device *bdev, struct iov_iter *iter,
-			     get_block_t get_block,
-			     dio_iodone_t end_io, dio_submit_t submit_io,
-			     int flags)
-{
-	/*
-	 * The block device state is needed in the end to finally
-	 * submit everything.  Since it's likely to be cache cold
-	 * prefetch it here as first thing to hide some of the
-	 * latency.
-	 *
-	 * Attempt to prefetch the pieces we likely need later.
-	 */
-	prefetch(&bdev->bd_disk->part_tbl);
-	prefetch(bdev->bd_disk->queue);
-	prefetch((char *)bdev->bd_disk->queue + SMP_CACHE_BYTES);
-
-	return do_blockdev_direct_IO(iocb, inode, bdev, iter, get_block,
-				     end_io, submit_io, flags);
-}
-
 EXPORT_SYMBOL(__blockdev_direct_IO);
 
 static __init int dio_init(void)
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel

[Ocfs2-devel] [PATCH 16/27] block: use bdev_alignment_offset in part_alignment_offset_show

[Christoph Hellwig via Ocfs2-devel]

Replace the open coded offset calculation with the proper helper.
This is an ABI change in that the -1 for a misaligned partition is
properly propagated, which can be considered a bug fix and matches
what is done on the whole device.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 block/partitions/core.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/block/partitions/core.c b/block/partitions/core.c
index 2ef8dfa1e5c85..240b3fff521e4 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -200,11 +200,7 @@ static ssize_t part_ro_show(struct device *dev,
 static ssize_t part_alignment_offset_show(struct device *dev,
 					  struct device_attribute *attr, char *buf)
 {
-	struct block_device *bdev = dev_to_bdev(dev);
-
-	return sprintf(buf, "%u\n",
-		queue_limit_alignment_offset(&bdev_get_queue(bdev)->limits,
-				bdev->bd_start_sect));
+	return sprintf(buf, "%u\n", bdev_alignment_offset(dev_to_bdev(dev)));
 }
 
 static ssize_t part_discard_alignment_show(struct device *dev,
-- 
2.30.2


_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel