Merge branch 'for-4.4/reservations' of git://git.kernel.dk/linux-block

Pull block reservation support from Jens Axboe:
 "This adds support for persistent reservations, both at the core level,
  as well as for sd and NVMe"

[ Background from the docs: "Persistent Reservations allow restricting
  access to block devices to specific initiators in a shared storage
  setup.  All implementations are expected to ensure the reservations
  survive a power loss and cover all connections in a multi path
  environment" ]

* 'for-4.4/reservations' of git://git.kernel.dk/linux-block:
  NVMe: Precedence error in nvme_pr_clear()
  nvme: add missing endianess annotations in nvme_pr_command
  NVMe: Add persistent reservation ops
  sd: implement the Persistent Reservation API
  block: add an API for Persistent Reservations
  block: cleanup blkdev_ioctl

Documentation/block/pr.txt

@@ -0,0 +1,119 @@
+
+Block layer support for Persistent Reservations
+===============================================
+
+The Linux kernel supports a user space interface for simplified
+Persistent Reservations which map to block devices that support
+these (like SCSI). Persistent Reservations allow restricting
+access to block devices to specific initiators in a shared storage
+setup.
+
+This document gives a general overview of the support ioctl commands.
+For a more detailed reference please refer the the SCSI Primary
+Commands standard, specifically the section on Reservations and the
+"PERSISTENT RESERVE IN" and "PERSISTENT RESERVE OUT" commands.
+
+All implementations are expected to ensure the reservations survive
+a power loss and cover all connections in a multi path environment.
+These behaviors are optional in SPC but will be automatically applied
+by Linux.
+
+
+The following types of reservations are supported:
+--------------------------------------------------
+
+ - PR_WRITE_EXCLUSIVE
+
+	Only the initiator that owns the reservation can write to the
+	device.  Any initiator can read from the device.
+
+ - PR_EXCLUSIVE_ACCESS
+
+	Only the initiator that owns the reservation can access the
+	device.
+
+ - PR_WRITE_EXCLUSIVE_REG_ONLY
+
+	Only initiators with a registered key can write to the device,
+	Any initiator can read from the device.
+
+ - PR_EXCLUSIVE_ACCESS_REG_ONLY
+
+	Only initiators with a registered key can access the device.
+
+ - PR_WRITE_EXCLUSIVE_ALL_REGS
+
+	Only initiators with a registered key can write to the device,
+	Any initiator can read from the device.
+	All initiators with a registered key are considered reservation
+	holders.
+	Please reference the SPC spec on the meaning of a reservation
+	holder if you want to use this type. 
+
+ - PR_EXCLUSIVE_ACCESS_ALL_REGS
+
+	Only initiators with a registered key can access the device.
+	All initiators with a registered key are considered reservation
+	holders.
+	Please reference the SPC spec on the meaning of a reservation
+	holder if you want to use this type. 
+
+
+The following ioctl are supported:
+----------------------------------
+
+1. IOC_PR_REGISTER
+
+This ioctl command registers a new reservation if the new_key argument
+is non-null.  If no existing reservation exists old_key must be zero,
+if an existing reservation should be replaced old_key must contain
+the old reservation key.
+
+If the new_key argument is 0 it unregisters the existing reservation passed
+in old_key.
+
+
+2. IOC_PR_RESERVE
+
+This ioctl command reserves the device and thus restricts access for other
+devices based on the type argument.  The key argument must be the existing
+reservation key for the device as acquired by the IOC_PR_REGISTER,
+IOC_PR_REGISTER_IGNORE, IOC_PR_PREEMPT or IOC_PR_PREEMPT_ABORT commands.
+
+
+3. IOC_PR_RELEASE
+
+This ioctl command releases the reservation specified by key and flags
+and thus removes any access restriction implied by it.
+
+
+4. IOC_PR_PREEMPT
+
+This ioctl command releases the existing reservation referred to by
+old_key and replaces it with a a new reservation of type for the
+reservation key new_key.
+
+
+5. IOC_PR_PREEMPT_ABORT
+
+This ioctl command works like IOC_PR_PREEMPT except that it also aborts
+any outstanding command sent over a connection identified by old_key.
+
+6. IOC_PR_CLEAR
+
+This ioctl command unregisters both key and any other reservation key
+registered with the device and drops any existing reservation.
+
+
+Flags
+-----
+
+All the ioctls have a flag field.  Currently only one flag is supported:
+
+ - PR_FL_IGNORE_KEY
+
+	Ignore the existing reservation key.  This is commonly supported for
+	IOC_PR_REGISTER, and some implementation may support the flag for
+	IOC_PR_RESERVE.
+
+For all unknown flags the kernel will return -EOPNOTSUPP.

block/ioctl.c

@@ -7,6 +7,7 @@
 #include <linux/backing-dev.h>
 #include <linux/fs.h>
 #include <linux/blktrace_api.h>
+#include <linux/pr.h>
 #include <asm/uaccess.h>
 
 static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user *arg)
@@ -193,10 +194,20 @@ int blkdev_reread_part(struct block_device *bdev)
 }
 EXPORT_SYMBOL(blkdev_reread_part);
 
-static int blk_ioctl_discard(struct block_device *bdev, uint64_t start,
+static int blk_ioctl_discard(struct block_device *bdev, fmode_t mode,
-			     uint64_t len, int secure)
+		unsigned long arg, unsigned long flags)
 {
-	unsigned long flags = 0;
+	uint64_t range[2];
+	uint64_t start, len;
+
+	if (!(mode & FMODE_WRITE))
+		return -EBADF;
+
+	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
+		return -EFAULT;
+
+	start = range[0];
+	len = range[1];
 
 	if (start & 511)
 		return -EINVAL;
@@ -207,14 +218,24 @@ static int blk_ioctl_discard(struct block_device *bdev, uint64_t start,
 
 	if (start + len > (i_size_read(bdev->bd_inode) >> 9))
 		return -EINVAL;
-	if (secure)
-		flags |= BLKDEV_DISCARD_SECURE;
 	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL, flags);
 }
 
-static int blk_ioctl_zeroout(struct block_device *bdev, uint64_t start,
+static int blk_ioctl_zeroout(struct block_device *bdev, fmode_t mode,
-			     uint64_t len)
+		unsigned long arg)
 {
+	uint64_t range[2];
+	uint64_t start, len;
+
+	if (!(mode & FMODE_WRITE))
+		return -EBADF;
+
+	if (copy_from_user(range, (void __user *)arg, sizeof(range)))
+		return -EFAULT;
+
+	start = range[0];
+	len = range[1];
+
 	if (start & 511)
 		return -EINVAL;
 	if (len & 511)
@@ -275,6 +296,96 @@ int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
  */
 EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);
 
+static int blkdev_pr_register(struct block_device *bdev,
+		struct pr_registration __user *arg)
+{
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	struct pr_registration reg;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+	if (!ops || !ops->pr_register)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&reg, arg, sizeof(reg)))
+		return -EFAULT;
+
+	if (reg.flags & ~PR_FL_IGNORE_KEY)
+		return -EOPNOTSUPP;
+	return ops->pr_register(bdev, reg.old_key, reg.new_key, reg.flags);
+}
+
+static int blkdev_pr_reserve(struct block_device *bdev,
+		struct pr_reservation __user *arg)
+{
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	struct pr_reservation rsv;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+	if (!ops || !ops->pr_reserve)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&rsv, arg, sizeof(rsv)))
+		return -EFAULT;
+
+	if (rsv.flags & ~PR_FL_IGNORE_KEY)
+		return -EOPNOTSUPP;
+	return ops->pr_reserve(bdev, rsv.key, rsv.type, rsv.flags);
+}
+
+static int blkdev_pr_release(struct block_device *bdev,
+		struct pr_reservation __user *arg)
+{
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	struct pr_reservation rsv;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+	if (!ops || !ops->pr_release)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&rsv, arg, sizeof(rsv)))
+		return -EFAULT;
+
+	if (rsv.flags)
+		return -EOPNOTSUPP;
+	return ops->pr_release(bdev, rsv.key, rsv.type);
+}
+
+static int blkdev_pr_preempt(struct block_device *bdev,
+		struct pr_preempt __user *arg, bool abort)
+{
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	struct pr_preempt p;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+	if (!ops || !ops->pr_preempt)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&p, arg, sizeof(p)))
+		return -EFAULT;
+
+	if (p.flags)
+		return -EOPNOTSUPP;
+	return ops->pr_preempt(bdev, p.old_key, p.new_key, p.type, abort);
+}
+
+static int blkdev_pr_clear(struct block_device *bdev,
+		struct pr_clear __user *arg)
+{
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	struct pr_clear c;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+	if (!ops || !ops->pr_clear)
+		return -EOPNOTSUPP;
+	if (copy_from_user(&c, arg, sizeof(c)))
+		return -EFAULT;
+
+	if (c.flags)
+		return -EOPNOTSUPP;
+	return ops->pr_clear(bdev, c.key);
+}
+
 /*
  * Is it an unrecognized ioctl? The correct returns are either
  * ENOTTY (final) or ENOIOCTLCMD ("I don't know this one, try a
@@ -295,89 +406,115 @@ static inline int is_unrecognized_ioctl(int ret)
 		ret == -ENOIOCTLCMD;
 }
 
+static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
+		unsigned cmd, unsigned long arg)
+{
+	int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
+	if (!is_unrecognized_ioctl(ret))
+		return ret;
+
+	fsync_bdev(bdev);
+	invalidate_bdev(bdev);
+	return 0;
+}
+
+static int blkdev_roset(struct block_device *bdev, fmode_t mode,
+		unsigned cmd, unsigned long arg)
+{
+	int ret, n;
+
+	ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
+	if (!is_unrecognized_ioctl(ret))
+		return ret;
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (get_user(n, (int __user *)arg))
+		return -EFAULT;
+	set_device_ro(bdev, n);
+	return 0;
+}
+
+static int blkdev_getgeo(struct block_device *bdev,
+		struct hd_geometry __user *argp)
+{
+	struct gendisk *disk = bdev->bd_disk;
+	struct hd_geometry geo;
+	int ret;
+
+	if (!argp)
+		return -EINVAL;
+	if (!disk->fops->getgeo)
+		return -ENOTTY;
+
+	/*
+	 * We need to set the startsect first, the driver may
+	 * want to override it.
+	 */
+	memset(&geo, 0, sizeof(geo));
+	geo.start = get_start_sect(bdev);
+	ret = disk->fops->getgeo(bdev, &geo);
+	if (ret)
+		return ret;
+	if (copy_to_user(argp, &geo, sizeof(geo)))
+		return -EFAULT;
+	return 0;
+}
+
+/* set the logical block size */
+static int blkdev_bszset(struct block_device *bdev, fmode_t mode,
+		int __user *argp)
+{
+	int ret, n;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (!argp)
+		return -EINVAL;
+	if (get_user(n, argp))
+		return -EFAULT;
+
+	if (!(mode & FMODE_EXCL)) {
+		bdgrab(bdev);
+		if (blkdev_get(bdev, mode | FMODE_EXCL, &bdev) < 0)
+			return -EBUSY;
+	}
+
+	ret = set_blocksize(bdev, n);
+	if (!(mode & FMODE_EXCL))
+		blkdev_put(bdev, mode | FMODE_EXCL);
+	return ret;
+}
+
 /*
  * always keep this in sync with compat_blkdev_ioctl()
  */
 int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 			unsigned long arg)
 {
-	struct gendisk *disk = bdev->bd_disk;
 	struct backing_dev_info *bdi;
+	void __user *argp = (void __user *)arg;
 	loff_t size;
-	int ret, n;
 	unsigned int max_sectors;
 
-	switch(cmd) {
+	switch (cmd) {
 	case BLKFLSBUF:
-		if (!capable(CAP_SYS_ADMIN))
+		return blkdev_flushbuf(bdev, mode, cmd, arg);
-			return -EACCES;
-
-		ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
-		if (!is_unrecognized_ioctl(ret))
-			return ret;
-
-		fsync_bdev(bdev);
-		invalidate_bdev(bdev);
-		return 0;
-
 	case BLKROSET:
-		ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
+		return blkdev_roset(bdev, mode, cmd, arg);
-		if (!is_unrecognized_ioctl(ret))
-			return ret;
-		if (!capable(CAP_SYS_ADMIN))
-			return -EACCES;
-		if (get_user(n, (int __user *)(arg)))
-			return -EFAULT;
-		set_device_ro(bdev, n);
-		return 0;
-
 	case BLKDISCARD:
-	case BLKSECDISCARD: {
+		return blk_ioctl_discard(bdev, mode, arg, 0);
-		uint64_t range[2];
+	case BLKSECDISCARD:
-
+		return blk_ioctl_discard(bdev, mode, arg,
-		if (!(mode & FMODE_WRITE))
+				BLKDEV_DISCARD_SECURE);
-			return -EBADF;
+	case BLKZEROOUT:
-
+		return blk_ioctl_zeroout(bdev, mode, arg);
-		if (copy_from_user(range, (void __user *)arg, sizeof(range)))
+	case HDIO_GETGEO:
-			return -EFAULT;
+		return blkdev_getgeo(bdev, argp);
-
-		return blk_ioctl_discard(bdev, range[0], range[1],
-					 cmd == BLKSECDISCARD);
-	}
-	case BLKZEROOUT: {
-		uint64_t range[2];
-
-		if (!(mode & FMODE_WRITE))
-			return -EBADF;
-
-		if (copy_from_user(range, (void __user *)arg, sizeof(range)))
-			return -EFAULT;
-
-		return blk_ioctl_zeroout(bdev, range[0], range[1]);
-	}
-
-	case HDIO_GETGEO: {
-		struct hd_geometry geo;
-
-		if (!arg)
-			return -EINVAL;
-		if (!disk->fops->getgeo)
-			return -ENOTTY;
-
-		/*
-		 * We need to set the startsect first, the driver may
-		 * want to override it.
-		 */
-		memset(&geo, 0, sizeof(geo));
-		geo.start = get_start_sect(bdev);
-		ret = disk->fops->getgeo(bdev, &geo);
-		if (ret)
-			return ret;
-		if (copy_to_user((struct hd_geometry __user *)arg, &geo,
-					sizeof(geo)))
-			return -EFAULT;
-		return 0;
-	}
 	case BLKRAGET:
 	case BLKFRAGET:
 		if (!arg)
@@ -414,28 +551,11 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 		bdi->ra_pages = (arg * 512) / PAGE_CACHE_SIZE;
 		return 0;
 	case BLKBSZSET:
-		/* set the logical block size */
+		return blkdev_bszset(bdev, mode, argp);
-		if (!capable(CAP_SYS_ADMIN))
-			return -EACCES;
-		if (!arg)
-			return -EINVAL;
-		if (get_user(n, (int __user *) arg))
-			return -EFAULT;
-		if (!(mode & FMODE_EXCL)) {
-			bdgrab(bdev);
-			if (blkdev_get(bdev, mode | FMODE_EXCL, &bdev) < 0)
-				return -EBUSY;
-		}
-		ret = set_blocksize(bdev, n);
-		if (!(mode & FMODE_EXCL))
-			blkdev_put(bdev, mode | FMODE_EXCL);
-		return ret;
 	case BLKPG:
-		ret = blkpg_ioctl(bdev, (struct blkpg_ioctl_arg __user *) arg);
+		return blkpg_ioctl(bdev, argp);
-		break;
 	case BLKRRPART:
-		ret = blkdev_reread_part(bdev);
+		return blkdev_reread_part(bdev);
-		break;
 	case BLKGETSIZE:
 		size = i_size_read(bdev->bd_inode);
 		if ((size >> 9) > ~0UL)
@@ -447,11 +567,21 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 	case BLKTRACESTOP:
 	case BLKTRACESETUP:
 	case BLKTRACETEARDOWN:
-		ret = blk_trace_ioctl(bdev, cmd, (char __user *) arg);
+		return blk_trace_ioctl(bdev, cmd, argp);
-		break;
+	case IOC_PR_REGISTER:
+		return blkdev_pr_register(bdev, argp);
+	case IOC_PR_RESERVE:
+		return blkdev_pr_reserve(bdev, argp);
+	case IOC_PR_RELEASE:
+		return blkdev_pr_release(bdev, argp);
+	case IOC_PR_PREEMPT:
+		return blkdev_pr_preempt(bdev, argp, false);
+	case IOC_PR_PREEMPT_ABORT:
+		return blkdev_pr_preempt(bdev, argp, true);
+	case IOC_PR_CLEAR:
+		return blkdev_pr_clear(bdev, argp);
 	default:
-		ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);
+		return __blkdev_driver_ioctl(bdev, mode, cmd, arg);
 	}
-	return ret;
 }
 EXPORT_SYMBOL_GPL(blkdev_ioctl);

drivers/nvme/host/pci.c

@@ -39,8 +39,10 @@
 #include <linux/slab.h>
 #include <linux/t10-pi.h>
 #include <linux/types.h>
+#include <linux/pr.h>
 #include <scsi/sg.h>
 #include <asm-generic/io-64-nonatomic-lo-hi.h>
+#include <asm/unaligned.h>
 
 #include <uapi/linux/nvme_ioctl.h>
 #include "nvme.h"
@@ -2064,6 +2066,98 @@ static int nvme_revalidate_disk(struct gendisk *disk)
 	return 0;
 }
 
+static char nvme_pr_type(enum pr_type type)
+{
+	switch (type) {
+	case PR_WRITE_EXCLUSIVE:
+		return 1;
+	case PR_EXCLUSIVE_ACCESS:
+		return 2;
+	case PR_WRITE_EXCLUSIVE_REG_ONLY:
+		return 3;
+	case PR_EXCLUSIVE_ACCESS_REG_ONLY:
+		return 4;
+	case PR_WRITE_EXCLUSIVE_ALL_REGS:
+		return 5;
+	case PR_EXCLUSIVE_ACCESS_ALL_REGS:
+		return 6;
+	default:
+		return 0;
+	}
+};
+
+static int nvme_pr_command(struct block_device *bdev, u32 cdw10,
+				u64 key, u64 sa_key, u8 op)
+{
+	struct nvme_ns *ns = bdev->bd_disk->private_data;
+	struct nvme_command c;
+	u8 data[16] = { 0, };
+
+	put_unaligned_le64(key, &data[0]);
+	put_unaligned_le64(sa_key, &data[8]);
+
+	memset(&c, 0, sizeof(c));
+	c.common.opcode = op;
+	c.common.nsid = cpu_to_le32(ns->ns_id);
+	c.common.cdw10[0] = cpu_to_le32(cdw10);
+
+	return nvme_submit_sync_cmd(ns->queue, &c, data, 16);
+}
+
+static int nvme_pr_register(struct block_device *bdev, u64 old,
+		u64 new, unsigned flags)
+{
+	u32 cdw10;
+
+	if (flags & ~PR_FL_IGNORE_KEY)
+		return -EOPNOTSUPP;
+
+	cdw10 = old ? 2 : 0;
+	cdw10 |= (flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0;
+	cdw10 |= (1 << 30) | (1 << 31); /* PTPL=1 */
+	return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_register);
+}
+
+static int nvme_pr_reserve(struct block_device *bdev, u64 key,
+		enum pr_type type, unsigned flags)
+{
+	u32 cdw10;
+
+	if (flags & ~PR_FL_IGNORE_KEY)
+		return -EOPNOTSUPP;
+
+	cdw10 = nvme_pr_type(type) << 8;
+	cdw10 |= ((flags & PR_FL_IGNORE_KEY) ? 1 << 3 : 0);
+	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_acquire);
+}
+
+static int nvme_pr_preempt(struct block_device *bdev, u64 old, u64 new,
+		enum pr_type type, bool abort)
+{
+	u32 cdw10 = nvme_pr_type(type) << 8 | abort ? 2 : 1;
+	return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_acquire);
+}
+
+static int nvme_pr_clear(struct block_device *bdev, u64 key)
+{
+	u32 cdw10 = 1 | (key ? 1 << 3 : 0);
+	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_register);
+}
+
+static int nvme_pr_release(struct block_device *bdev, u64 key, enum pr_type type)
+{
+	u32 cdw10 = nvme_pr_type(type) << 8 | key ? 1 << 3 : 0;
+	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_release);
+}
+
+static const struct pr_ops nvme_pr_ops = {
+	.pr_register	= nvme_pr_register,
+	.pr_reserve	= nvme_pr_reserve,
+	.pr_release	= nvme_pr_release,
+	.pr_preempt	= nvme_pr_preempt,
+	.pr_clear	= nvme_pr_clear,
+};
+
 static const struct block_device_operations nvme_fops = {
 	.owner		= THIS_MODULE,
 	.ioctl		= nvme_ioctl,
@@ -2072,6 +2166,7 @@ static const struct block_device_operations nvme_fops = {
 	.release	= nvme_release,
 	.getgeo		= nvme_getgeo,
 	.revalidate_disk= nvme_revalidate_disk,
+	.pr_ops		= &nvme_pr_ops,
 };
 
 static int nvme_kthread(void *data)

drivers/scsi/sd.c

@@ -51,6 +51,7 @@
 #include <linux/async.h>
 #include <linux/slab.h>
 #include <linux/pm_runtime.h>
+#include <linux/pr.h>
 #include <asm/uaccess.h>
 #include <asm/unaligned.h>
 
@@ -1535,6 +1536,100 @@ static int sd_compat_ioctl(struct block_device *bdev, fmode_t mode,
 }
 #endif
 
+static char sd_pr_type(enum pr_type type)
+{
+	switch (type) {
+	case PR_WRITE_EXCLUSIVE:
+		return 0x01;
+	case PR_EXCLUSIVE_ACCESS:
+		return 0x03;
+	case PR_WRITE_EXCLUSIVE_REG_ONLY:
+		return 0x05;
+	case PR_EXCLUSIVE_ACCESS_REG_ONLY:
+		return 0x06;
+	case PR_WRITE_EXCLUSIVE_ALL_REGS:
+		return 0x07;
+	case PR_EXCLUSIVE_ACCESS_ALL_REGS:
+		return 0x08;
+	default:
+		return 0;
+	}
+};
+
+static int sd_pr_command(struct block_device *bdev, u8 sa,
+		u64 key, u64 sa_key, u8 type, u8 flags)
+{
+	struct scsi_device *sdev = scsi_disk(bdev->bd_disk)->device;
+	struct scsi_sense_hdr sshdr;
+	int result;
+	u8 cmd[16] = { 0, };
+	u8 data[24] = { 0, };
+
+	cmd[0] = PERSISTENT_RESERVE_OUT;
+	cmd[1] = sa;
+	cmd[2] = type;
+	put_unaligned_be32(sizeof(data), &cmd[5]);
+
+	put_unaligned_be64(key, &data[0]);
+	put_unaligned_be64(sa_key, &data[8]);
+	data[20] = flags;
+
+	result = scsi_execute_req(sdev, cmd, DMA_TO_DEVICE, &data, sizeof(data),
+			&sshdr, SD_TIMEOUT, SD_MAX_RETRIES, NULL);
+
+	if ((driver_byte(result) & DRIVER_SENSE) &&
+	    (scsi_sense_valid(&sshdr))) {
+		sdev_printk(KERN_INFO, sdev, "PR command failed: %d\n", result);
+		scsi_print_sense_hdr(sdev, NULL, &sshdr);
+	}
+
+	return result;
+}
+
+static int sd_pr_register(struct block_device *bdev, u64 old_key, u64 new_key,
+		u32 flags)
+{
+	if (flags & ~PR_FL_IGNORE_KEY)
+		return -EOPNOTSUPP;
+	return sd_pr_command(bdev, (flags & PR_FL_IGNORE_KEY) ? 0x06 : 0x00,
+			old_key, new_key, 0,
+			(1 << 0) /* APTPL */ |
+			(1 << 2) /* ALL_TG_PT */);
+}
+
+static int sd_pr_reserve(struct block_device *bdev, u64 key, enum pr_type type,
+		u32 flags)
+{
+	if (flags)
+		return -EOPNOTSUPP;
+	return sd_pr_command(bdev, 0x01, key, 0, sd_pr_type(type), 0);
+}
+
+static int sd_pr_release(struct block_device *bdev, u64 key, enum pr_type type)
+{
+	return sd_pr_command(bdev, 0x02, key, 0, sd_pr_type(type), 0);
+}
+
+static int sd_pr_preempt(struct block_device *bdev, u64 old_key, u64 new_key,
+		enum pr_type type, bool abort)
+{
+	return sd_pr_command(bdev, abort ? 0x05 : 0x04, old_key, new_key,
+			     sd_pr_type(type), 0);
+}
+
+static int sd_pr_clear(struct block_device *bdev, u64 key)
+{
+	return sd_pr_command(bdev, 0x03, key, 0, 0, 0);
+}
+
+static const struct pr_ops sd_pr_ops = {
+	.pr_register	= sd_pr_register,
+	.pr_reserve	= sd_pr_reserve,
+	.pr_release	= sd_pr_release,
+	.pr_preempt	= sd_pr_preempt,
+	.pr_clear	= sd_pr_clear,
+};
+
 static const struct block_device_operations sd_fops = {
 	.owner			= THIS_MODULE,
 	.open			= sd_open,
@@ -1547,6 +1642,7 @@ static const struct block_device_operations sd_fops = {
 	.check_events		= sd_check_events,
 	.revalidate_disk	= sd_revalidate_disk,
 	.unlock_native_capacity	= sd_unlock_native_capacity,
+	.pr_ops			= &sd_pr_ops,
 };
 
 /**

include/linux/blkdev.h

@@ -35,6 +35,7 @@ struct sg_io_hdr;
 struct bsg_job;
 struct blkcg_gq;
 struct blk_flush_queue;
+struct pr_ops;
 
 #define BLKDEV_MIN_RQ	4
 #define BLKDEV_MAX_RQ	128	/* Default maximum */
@@ -1629,6 +1630,7 @@ struct block_device_operations {
 	/* this callback is with swap_lock and sometimes page table lock held */
 	void (*swap_slot_free_notify) (struct block_device *, unsigned long);
 	struct module *owner;
+	const struct pr_ops *pr_ops;
 };
 
 extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int,

include/linux/pr.h

@@ -0,0 +1,18 @@
+#ifndef LINUX_PR_H
+#define LINUX_PR_H
+
+#include <uapi/linux/pr.h>
+
+struct pr_ops {
+	int (*pr_register)(struct block_device *bdev, u64 old_key, u64 new_key,
+			u32 flags);
+	int (*pr_reserve)(struct block_device *bdev, u64 key,
+			enum pr_type type, u32 flags);
+	int (*pr_release)(struct block_device *bdev, u64 key,
+			enum pr_type type);
+	int (*pr_preempt)(struct block_device *bdev, u64 old_key, u64 new_key,
+			enum pr_type type, bool abort);
+	int (*pr_clear)(struct block_device *bdev, u64 key);
+};
+
+#endif /* LINUX_PR_H */

include/uapi/linux/pr.h

@@ -0,0 +1,48 @@
+#ifndef _UAPI_PR_H
+#define _UAPI_PR_H
+
+enum pr_type {
+	PR_WRITE_EXCLUSIVE		= 1,
+	PR_EXCLUSIVE_ACCESS		= 2,
+	PR_WRITE_EXCLUSIVE_REG_ONLY	= 3,
+	PR_EXCLUSIVE_ACCESS_REG_ONLY	= 4,
+	PR_WRITE_EXCLUSIVE_ALL_REGS	= 5,
+	PR_EXCLUSIVE_ACCESS_ALL_REGS	= 6,
+};
+
+struct pr_reservation {
+	__u64	key;
+	__u32	type;
+	__u32	flags;
+};
+
+struct pr_registration {
+	__u64	old_key;
+	__u64	new_key;
+	__u32	flags;
+	__u32	__pad;
+};
+
+struct pr_preempt {
+	__u64	old_key;
+	__u64	new_key;
+	__u32	type;
+	__u32	flags;
+};
+
+struct pr_clear {
+	__u64	key;
+	__u32	flags;
+	__u32	__pad;
+};
+
+#define PR_FL_IGNORE_KEY	(1 << 0)	/* ignore existing key */
+
+#define IOC_PR_REGISTER		_IOW('p', 200, struct pr_registration)
+#define IOC_PR_RESERVE		_IOW('p', 201, struct pr_reservation)
+#define IOC_PR_RELEASE		_IOW('p', 202, struct pr_reservation)
+#define IOC_PR_PREEMPT		_IOW('p', 203, struct pr_preempt)
+#define IOC_PR_PREEMPT_ABORT	_IOW('p', 204, struct pr_preempt)
+#define IOC_PR_CLEAR		_IOW('p', 205, struct pr_clear)
+
+#endif /* _UAPI_PR_H */