1 files changed, 145 insertions, 206 deletions

diff --git a/drivers/dax/super.c b/drivers/dax/super.c
index b882cf8106ea..da4438f3188c 100644
--- a/drivers/dax/super.c
+++ b/drivers/dax/super.c
@@ -7,10 +7,8 @@
 #include <linux/mount.h>
 #include <linux/pseudo_fs.h>
 #include <linux/magic.h>
-#include <linux/genhd.h>
 #include <linux/pfn_t.h>
 #include <linux/cdev.h>
-#include <linux/hash.h>
 #include <linux/slab.h>
 #include <linux/uio.h>
 #include <linux/dax.h>
@@ -21,18 +19,20 @@
  * struct dax_device - anchor object for dax services
  * @inode: core vfs
  * @cdev: optional character interface for "device dax"
- * @host: optional name for lookups where the device path is not available
  * @private: dax driver private data
  * @flags: state and boolean properties
+ * @ops: operations for this device
+ * @holder_data: holder of a dax_device: could be filesystem or mapped device
+ * @holder_ops: operations for the inner holder
  */
 struct dax_device {
-	struct hlist_node list;
 	struct inode inode;
 	struct cdev cdev;
-	const char *host;
 	void *private;
 	unsigned long flags;
 	const struct dax_operations *ops;
+	void *holder_data;
+	const struct dax_holder_operations *holder_ops;
 };
 
 static dev_t dax_devt;
@@ -42,10 +42,6 @@ static DEFINE_IDA(dax_minor_ida);
 static struct kmem_cache *dax_cache __read_mostly;
 static struct super_block *dax_superblock __read_mostly;
 
-#define DAX_HASH_SIZE (PAGE_SIZE / sizeof(struct hlist_head))
-static struct hlist_head dax_host_list[DAX_HASH_SIZE];
-static DEFINE_SPINLOCK(dax_host_lock);
-
 int dax_read_lock(void)
 {
 	return srcu_read_lock(&dax_srcu);
@@ -58,169 +54,72 @@ void dax_read_unlock(int id)
 }
 EXPORT_SYMBOL_GPL(dax_read_unlock);
 
-static int dax_host_hash(const char *host)
-{
-	return hashlen_hash(hashlen_string("DAX", host)) % DAX_HASH_SIZE;
-}
-
-#ifdef CONFIG_BLOCK
+#if defined(CONFIG_BLOCK) && defined(CONFIG_FS_DAX)
 #include <linux/blkdev.h>
 
-int bdev_dax_pgoff(struct block_device *bdev, sector_t sector, size_t size,
-		pgoff_t *pgoff)
+static DEFINE_XARRAY(dax_hosts);
+
+int dax_add_host(struct dax_device *dax_dev, struct gendisk *disk)
 {
-	sector_t start_sect = bdev ? get_start_sect(bdev) : 0;
-	phys_addr_t phys_off = (start_sect + sector) * 512;
+	return xa_insert(&dax_hosts, (unsigned long)disk, dax_dev, GFP_KERNEL);
+}
+EXPORT_SYMBOL_GPL(dax_add_host);
 
-	if (pgoff)
-		*pgoff = PHYS_PFN(phys_off);
-	if (phys_off % PAGE_SIZE || size % PAGE_SIZE)
-		return -EINVAL;
-	return 0;
+void dax_remove_host(struct gendisk *disk)
+{
+	xa_erase(&dax_hosts, (unsigned long)disk);
 }
-EXPORT_SYMBOL(bdev_dax_pgoff);
+EXPORT_SYMBOL_GPL(dax_remove_host);
 
-#if IS_ENABLED(CONFIG_FS_DAX)
 /**
- * dax_get_by_host() - temporary lookup mechanism for filesystem-dax
- * @host: alternate name for the device registered by a dax driver
+ * fs_dax_get_by_bdev() - temporary lookup mechanism for filesystem-dax
+ * @bdev: block device to find a dax_device for
+ * @start_off: returns the byte offset into the dax_device that @bdev starts
+ * @holder: filesystem or mapped device inside the dax_device
+ * @ops: operations for the inner holder
  */
-static struct dax_device *dax_get_by_host(const char *host)
+struct dax_device *fs_dax_get_by_bdev(struct block_device *bdev, u64 *start_off,
+		void *holder, const struct dax_holder_operations *ops)
 {
-	struct dax_device *dax_dev, *found = NULL;
-	int hash, id;
-
-	if (!host)
-		return NULL;
-
-	hash = dax_host_hash(host);
-
-	id = dax_read_lock();
-	spin_lock(&dax_host_lock);
-	hlist_for_each_entry(dax_dev, &dax_host_list[hash], list) {
-		if (!dax_alive(dax_dev)
-				|| strcmp(host, dax_dev->host) != 0)
-			continue;
-
-		if (igrab(&dax_dev->inode))
-			found = dax_dev;
-		break;
-	}
-	spin_unlock(&dax_host_lock);
-	dax_read_unlock(id);
-
-	return found;
-}
+	struct dax_device *dax_dev;
+	u64 part_size;
+	int id;
 
-struct dax_device *fs_dax_get_by_bdev(struct block_device *bdev)
-{
 	if (!blk_queue_dax(bdev->bd_disk->queue))
 		return NULL;
-	return dax_get_by_host(bdev->bd_disk->disk_name);
-}
-EXPORT_SYMBOL_GPL(fs_dax_get_by_bdev);
 
-bool generic_fsdax_supported(struct dax_device *dax_dev,
-		struct block_device *bdev, int blocksize, sector_t start,
-		sector_t sectors)
-{
-	bool dax_enabled = false;
-	pgoff_t pgoff, pgoff_end;
-	void *kaddr, *end_kaddr;
-	pfn_t pfn, end_pfn;
-	sector_t last_page;
-	long len, len2;
-	int err, id;
-
-	if (blocksize != PAGE_SIZE) {
-		pr_info("%pg: error: unsupported blocksize for dax\n", bdev);
-		return false;
-	}
-
-	if (!dax_dev) {
-		pr_debug("%pg: error: dax unsupported by block device\n", bdev);
-		return false;
-	}
-
-	err = bdev_dax_pgoff(bdev, start, PAGE_SIZE, &pgoff);
-	if (err) {
+	*start_off = get_start_sect(bdev) * SECTOR_SIZE;
+	part_size = bdev_nr_sectors(bdev) * SECTOR_SIZE;
+	if (*start_off % PAGE_SIZE || part_size % PAGE_SIZE) {
 		pr_info("%pg: error: unaligned partition for dax\n", bdev);
-		return false;
-	}
-
-	last_page = PFN_DOWN((start + sectors - 1) * 512) * PAGE_SIZE / 512;
-	err = bdev_dax_pgoff(bdev, last_page, PAGE_SIZE, &pgoff_end);
-	if (err) {
-		pr_info("%pg: error: unaligned partition for dax\n", bdev);
-		return false;
+		return NULL;
 	}
 
 	id = dax_read_lock();
-	len = dax_direct_access(dax_dev, pgoff, 1, &kaddr, &pfn);
-	len2 = dax_direct_access(dax_dev, pgoff_end, 1, &end_kaddr, &end_pfn);
-
-	if (len < 1 || len2 < 1) {
-		pr_info("%pg: error: dax access failed (%ld)\n",
-				bdev, len < 1 ? len : len2);
-		dax_read_unlock(id);
-		return false;
-	}
-
-	if (IS_ENABLED(CONFIG_FS_DAX_LIMITED) && pfn_t_special(pfn)) {
-		/*
-		 * An arch that has enabled the pmem api should also
-		 * have its drivers support pfn_t_devmap()
-		 *
-		 * This is a developer warning and should not trigger in
-		 * production. dax_flush() will crash since it depends
-		 * on being able to do (page_address(pfn_to_page())).
-		 */
-		WARN_ON(IS_ENABLED(CONFIG_ARCH_HAS_PMEM_API));
-		dax_enabled = true;
-	} else if (pfn_t_devmap(pfn) && pfn_t_devmap(end_pfn)) {
-		struct dev_pagemap *pgmap, *end_pgmap;
-
-		pgmap = get_dev_pagemap(pfn_t_to_pfn(pfn), NULL);
-		end_pgmap = get_dev_pagemap(pfn_t_to_pfn(end_pfn), NULL);
-		if (pgmap && pgmap == end_pgmap && pgmap->type == MEMORY_DEVICE_FS_DAX
-				&& pfn_t_to_page(pfn)->pgmap == pgmap
-				&& pfn_t_to_page(end_pfn)->pgmap == pgmap
-				&& pfn_t_to_pfn(pfn) == PHYS_PFN(__pa(kaddr))
-				&& pfn_t_to_pfn(end_pfn) == PHYS_PFN(__pa(end_kaddr)))
-			dax_enabled = true;
-		put_dev_pagemap(pgmap);
-		put_dev_pagemap(end_pgmap);
-
+	dax_dev = xa_load(&dax_hosts, (unsigned long)bdev->bd_disk);
+	if (!dax_dev || !dax_alive(dax_dev) || !igrab(&dax_dev->inode))
+		dax_dev = NULL;
+	else if (holder) {
+		if (!cmpxchg(&dax_dev->holder_data, NULL, holder))
+			dax_dev->holder_ops = ops;
+		else
+			dax_dev = NULL;
 	}
 	dax_read_unlock(id);
 
-	if (!dax_enabled) {
-		pr_info("%pg: error: dax support not enabled\n", bdev);
-		return false;
-	}
-	return true;
+	return dax_dev;
 }
-EXPORT_SYMBOL_GPL(generic_fsdax_supported);
+EXPORT_SYMBOL_GPL(fs_dax_get_by_bdev);
 
-bool dax_supported(struct dax_device *dax_dev, struct block_device *bdev,
-		int blocksize, sector_t start, sector_t len)
+void fs_put_dax(struct dax_device *dax_dev, void *holder)
 {
-	bool ret = false;
-	int id;
-
-	if (!dax_dev)
-		return false;
-
-	id = dax_read_lock();
-	if (dax_alive(dax_dev) && dax_dev->ops->dax_supported)
-		ret = dax_dev->ops->dax_supported(dax_dev, bdev, blocksize,
-						  start, len);
-	dax_read_unlock(id);
-	return ret;
+	if (dax_dev && holder &&
+	    cmpxchg(&dax_dev->holder_data, holder, NULL) == holder)
+		dax_dev->holder_ops = NULL;
+	put_dax(dax_dev);
 }
-EXPORT_SYMBOL_GPL(dax_supported);
-#endif /* CONFIG_FS_DAX */
-#endif /* CONFIG_BLOCK */
+EXPORT_SYMBOL_GPL(fs_put_dax);
+#endif /* CONFIG_BLOCK && CONFIG_FS_DAX */
 
 enum dax_device_flags {
 	/* !alive + rcu grace period == no new operations / mappings */
@@ -229,6 +128,10 @@ enum dax_device_flags {
 	DAXDEV_WRITE_CACHE,
 	/* flag to check if device supports synchronous flush */
 	DAXDEV_SYNC,
+	/* do not leave the caches dirty after writes */
+	DAXDEV_NOCACHE,
+	/* handle CPU fetch exceptions during reads */
+	DAXDEV_NOMC,
 };
 
 /**
@@ -236,6 +139,7 @@ enum dax_device_flags {
  * @dax_dev: a dax_device instance representing the logical memory range
  * @pgoff: offset in pages from the start of the device to translate
  * @nr_pages: number of consecutive pages caller can handle relative to @pfn
+ * @mode: indicator on normal access or recovery write
  * @kaddr: output parameter that returns a virtual address mapping of pfn
  * @pfn: output parameter that returns an absolute pfn translation of @pgoff
  *
@@ -243,7 +147,7 @@ enum dax_device_flags {
  * pages accessible at the device relative @pgoff.
  */
 long dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, long nr_pages,
-		void **kaddr, pfn_t *pfn)
+		enum dax_access_mode mode, void **kaddr, pfn_t *pfn)
 {
 	long avail;
 
@@ -257,7 +161,7 @@ long dax_direct_access(struct dax_device *dax_dev, pgoff_t pgoff, long nr_pages,
 		return -EINVAL;
 
 	avail = dax_dev->ops->direct_access(dax_dev, pgoff, nr_pages,
-			kaddr, pfn);
+			mode, kaddr, pfn);
 	if (!avail)
 		return -ERANGE;
 	return min(avail, nr_pages);
@@ -270,9 +174,15 @@ size_t dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr,
 	if (!dax_alive(dax_dev))
 		return 0;
 
-	return dax_dev->ops->copy_from_iter(dax_dev, pgoff, addr, bytes, i);
+	/*
+	 * The userspace address for the memory copy has already been validated
+	 * via access_ok() in vfs_write, so use the 'no check' version to bypass
+	 * the HARDENED_USERCOPY overhead.
+	 */
+	if (test_bit(DAXDEV_NOCACHE, &dax_dev->flags))
+		return _copy_from_iter_flushcache(addr, bytes, i);
+	return _copy_from_iter(addr, bytes, i);
 }
-EXPORT_SYMBOL_GPL(dax_copy_from_iter);
 
 size_t dax_copy_to_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr,
 		size_t bytes, struct iov_iter *i)
@@ -280,9 +190,15 @@ size_t dax_copy_to_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr,
 	if (!dax_alive(dax_dev))
 		return 0;
 
-	return dax_dev->ops->copy_to_iter(dax_dev, pgoff, addr, bytes, i);
+	/*
+	 * The userspace address for the memory copy has already been validated
+	 * via access_ok() in vfs_red, so use the 'no check' version to bypass
+	 * the HARDENED_USERCOPY overhead.
+	 */
+	if (test_bit(DAXDEV_NOMC, &dax_dev->flags))
+		return _copy_mc_to_iter(addr, bytes, i);
+	return _copy_to_iter(addr, bytes, i);
 }
-EXPORT_SYMBOL_GPL(dax_copy_to_iter);
 
 int dax_zero_page_range(struct dax_device *dax_dev, pgoff_t pgoff,
 			size_t nr_pages)
@@ -301,6 +217,38 @@ int dax_zero_page_range(struct dax_device *dax_dev, pgoff_t pgoff,
 }
 EXPORT_SYMBOL_GPL(dax_zero_page_range);
 
+size_t dax_recovery_write(struct dax_device *dax_dev, pgoff_t pgoff,
+		void *addr, size_t bytes, struct iov_iter *iter)
+{
+	if (!dax_dev->ops->recovery_write)
+		return 0;
+	return dax_dev->ops->recovery_write(dax_dev, pgoff, addr, bytes, iter);
+}
+EXPORT_SYMBOL_GPL(dax_recovery_write);
+
+int dax_holder_notify_failure(struct dax_device *dax_dev, u64 off,
+			      u64 len, int mf_flags)
+{
+	int rc, id;
+
+	id = dax_read_lock();
+	if (!dax_alive(dax_dev)) {
+		rc = -ENXIO;
+		goto out;
+	}
+
+	if (!dax_dev->holder_ops) {
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+
+	rc = dax_dev->holder_ops->notify_failure(dax_dev, off, len, mf_flags);
+out:
+	dax_read_unlock(id);
+	return rc;
+}
+EXPORT_SYMBOL_GPL(dax_holder_notify_failure);
+
 #ifdef CONFIG_ARCH_HAS_PMEM_API
 void arch_wb_cache_pmem(void *addr, size_t size);
 void dax_flush(struct dax_device *dax_dev, void *addr, size_t size)
@@ -332,17 +280,29 @@ bool dax_write_cache_enabled(struct dax_device *dax_dev)
 }
 EXPORT_SYMBOL_GPL(dax_write_cache_enabled);
 
-bool __dax_synchronous(struct dax_device *dax_dev)
+bool dax_synchronous(struct dax_device *dax_dev)
 {
 	return test_bit(DAXDEV_SYNC, &dax_dev->flags);
 }
-EXPORT_SYMBOL_GPL(__dax_synchronous);
+EXPORT_SYMBOL_GPL(dax_synchronous);
 
-void __set_dax_synchronous(struct dax_device *dax_dev)
+void set_dax_synchronous(struct dax_device *dax_dev)
 {
 	set_bit(DAXDEV_SYNC, &dax_dev->flags);
 }
-EXPORT_SYMBOL_GPL(__set_dax_synchronous);
+EXPORT_SYMBOL_GPL(set_dax_synchronous);
+
+void set_dax_nocache(struct dax_device *dax_dev)
+{
+	set_bit(DAXDEV_NOCACHE, &dax_dev->flags);
+}
+EXPORT_SYMBOL_GPL(set_dax_nocache);
+
+void set_dax_nomc(struct dax_device *dax_dev)
+{
+	set_bit(DAXDEV_NOMC, &dax_dev->flags);
+}
+EXPORT_SYMBOL_GPL(set_dax_nomc);
 
 bool dax_alive(struct dax_device *dax_dev)
 {
@@ -362,13 +322,15 @@ void kill_dax(struct dax_device *dax_dev)
 	if (!dax_dev)
 		return;
 
-	clear_bit(DAXDEV_ALIVE, &dax_dev->flags);
+	if (dax_dev->holder_data != NULL)
+		dax_holder_notify_failure(dax_dev, 0, U64_MAX, 0);
 
+	clear_bit(DAXDEV_ALIVE, &dax_dev->flags);
 	synchronize_srcu(&dax_srcu);
 
-	spin_lock(&dax_host_lock);
-	hlist_del_init(&dax_dev->list);
-	spin_unlock(&dax_host_lock);
+	/* clear holder data */
+	dax_dev->holder_ops = NULL;
+	dax_dev->holder_data = NULL;
 }
 EXPORT_SYMBOL_GPL(kill_dax);
 
@@ -383,7 +345,7 @@ static struct inode *dax_alloc_inode(struct super_block *sb)
 	struct dax_device *dax_dev;
 	struct inode *inode;
 
-	dax_dev = kmem_cache_alloc(dax_cache, GFP_KERNEL);
+	dax_dev = alloc_inode_sb(sb, dax_cache, GFP_KERNEL);
 	if (!dax_dev)
 		return NULL;
 
@@ -400,10 +362,8 @@ static struct dax_device *to_dax_dev(struct inode *inode)
 static void dax_free_inode(struct inode *inode)
 {
 	struct dax_device *dax_dev = to_dax_dev(inode);
-	kfree(dax_dev->host);
-	dax_dev->host = NULL;
 	if (inode->i_rdev)
-		ida_simple_remove(&dax_minor_ida, iminor(inode));
+		ida_free(&dax_minor_ida, iminor(inode));
 	kmem_cache_free(dax_cache, dax_dev);
 }
 
@@ -476,65 +436,30 @@ static struct dax_device *dax_dev_get(dev_t devt)
 	return dax_dev;
 }
 
-static void dax_add_host(struct dax_device *dax_dev, const char *host)
-{
-	int hash;
-
-	/*
-	 * Unconditionally init dax_dev since it's coming from a
-	 * non-zeroed slab cache
-	 */
-	INIT_HLIST_NODE(&dax_dev->list);
-	dax_dev->host = host;
-	if (!host)
-		return;
-
-	hash = dax_host_hash(host);
-	spin_lock(&dax_host_lock);
-	hlist_add_head(&dax_dev->list, &dax_host_list[hash]);
-	spin_unlock(&dax_host_lock);
-}
-
-struct dax_device *alloc_dax(void *private, const char *__host,
-		const struct dax_operations *ops, unsigned long flags)
+struct dax_device *alloc_dax(void *private, const struct dax_operations *ops)
 {
 	struct dax_device *dax_dev;
-	const char *host;
 	dev_t devt;
 	int minor;
 
-	if (ops && !ops->zero_page_range) {
-		pr_debug("%s: error: device does not provide dax"
-			 " operation zero_page_range()\n",
-			 __host ? __host : "Unknown");
+	if (WARN_ON_ONCE(ops && !ops->zero_page_range))
 		return ERR_PTR(-EINVAL);
-	}
-
-	host = kstrdup(__host, GFP_KERNEL);
-	if (__host && !host)
-		return ERR_PTR(-ENOMEM);
 
-	minor = ida_simple_get(&dax_minor_ida, 0, MINORMASK+1, GFP_KERNEL);
+	minor = ida_alloc_max(&dax_minor_ida, MINORMASK, GFP_KERNEL);
 	if (minor < 0)
-		goto err_minor;
+		return ERR_PTR(-ENOMEM);
 
 	devt = MKDEV(MAJOR(dax_devt), minor);
 	dax_dev = dax_dev_get(devt);
 	if (!dax_dev)
 		goto err_dev;
 
-	dax_add_host(dax_dev, host);
 	dax_dev->ops = ops;
 	dax_dev->private = private;
-	if (flags & DAXDEV_F_SYNC)
-		set_dax_synchronous(dax_dev);
-
 	return dax_dev;
 
  err_dev:
-	ida_simple_remove(&dax_minor_ida, minor);
- err_minor:
-	kfree(host);
+	ida_free(&dax_minor_ida, minor);
 	return ERR_PTR(-ENOMEM);
 }
 EXPORT_SYMBOL_GPL(alloc_dax);
@@ -548,6 +473,19 @@ void put_dax(struct dax_device *dax_dev)
 EXPORT_SYMBOL_GPL(put_dax);
 
 /**
+ * dax_holder() - obtain the holder of a dax device
+ * @dax_dev: a dax_device instance
+
+ * Return: the holder's data which represents the holder if registered,
+ * otherwize NULL.
+ */
+void *dax_holder(struct dax_device *dax_dev)
+{
+	return dax_dev->holder_data;
+}
+EXPORT_SYMBOL_GPL(dax_holder);
+
+/**
  * inode_dax: convert a public inode into its dax_dev
  * @inode: An inode with i_cdev pointing to a dax_dev
  *
@@ -614,6 +552,7 @@ static int dax_fs_init(void)
 static void dax_fs_exit(void)
 {
 	kern_unmount(dax_mnt);
+	rcu_barrier();
 	kmem_cache_destroy(dax_cache);
 }