Skip to contentSkip to conversation

block: Make blkdev_get_by_*() return handle

PATCH Applied Jan Kara 10 participants 109 messages Aug 23, 2023 – Oct 2, 2023

Subsystems (32)
lore
Share & tools
lore

Jan Kara

author reviewer
#0 lore raw

v3 cover block: Make blkdev_get_by_*() return handle

Hello,

this is a v3 of the patch series which implements the idea of blkdev_get_by_*() calls returning bdev_handle which is then passed to blkdev_put() [1]. This makes the get and put calls for bdevs more obviously matching and allows us to propagate context from get to put without having to modify all the users (again!). In particular I need to propagate used open flags to blkdev_put() to be able count writeable opens and add support for blocking writes to mounted block devices. I'll send that series separately.

The series is based on Christian's vfs tree as of today as there is quite some overlap. Patches have passed some reasonable testing - I've tested block changes, md, dm, bcache, xfs, btrfs, ext4, swap. More testing or review is always welcome. Thanks! I've pushed out the full branch to:

git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs.git bdev_handle

to ease review / testing. Since there were not many comments for v2 and Christoph has acked the series I think we should start discussing how to merge the series. Most collisions with this series seem to happen in the filesystems area so VFS tree would seem as the least painful way to merge this. Jens, are you OK with that?

Changes since v2:
* Rebased on top of current vfs tree
* Added some acks
* Reflected minor nits from Christoph
* Added missing conversion of blkdev_put() calls in cramfs and erofs
* Fixed possible leak of bdev handle in xfs if logdev is the same as fs dev

Changes since v1:
* Rebased on top of current vfs tree
* Renamed final functions to bdev_open_by_*() and bdev_release()
* Fixed detection of exclusive open in blkdev_ioctl() and blkdev_fallocate()
* Fixed swap conversion to properly reinitialize swap_info->bdev_handle
* Fixed xfs conversion to not oops with rtdev without logdev
* Couple other minor fixups

Honza

[1] https://lore.kernel.org/all/[email protected]

Previous versions:

author reviewer
#1 lore raw

PATCH v3 1/29 block: Provide bdev_open_* functions

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. This will eventually allow us to pass one more argument to blkdev_put() (renamed to bdev_release()) without too much hassle.

---
 block/bdev.c           | 48 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/blkdev.h | 10 +++++++++
 2 files changed, 58 insertions(+)
block/bdev.c +48 -0
index a20263fa27a4..fce150f9e66d 100644
@@ -829,6 +829,25 @@ struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 }
 EXPORT_SYMBOL(blkdev_get_by_dev);
 
+struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
+				     const struct blk_holder_ops *hops)
+{
+	struct bdev_handle *handle = kmalloc(sizeof(*handle), GFP_KERNEL);
+	struct block_device *bdev;
+
+	if (!handle)
+		return ERR_PTR(-ENOMEM);
+	bdev = blkdev_get_by_dev(dev, mode, holder, hops);
+	if (IS_ERR(bdev)) {
+		kfree(handle);
+		return ERR_CAST(bdev);
+	}
+	handle->bdev = bdev;
+	handle->holder = holder;
+	return handle;
+}
+EXPORT_SYMBOL(bdev_open_by_dev);
+
 /**
  * blkdev_get_by_path - open a block device by name
  * @path: path to the block device to open
@@ -867,6 +886,28 @@ struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
 }
 EXPORT_SYMBOL(blkdev_get_by_path);
 
+struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
+		void *holder, const struct blk_holder_ops *hops)
+{
+	struct bdev_handle *handle;
+	dev_t dev;
+	int error;
+
+	error = lookup_bdev(path, &dev);
+	if (error)
+		return ERR_PTR(error);
+
+	handle = bdev_open_by_dev(dev, mode, holder, hops);
+	if (!IS_ERR(handle) && (mode & BLK_OPEN_WRITE) &&
+	    bdev_read_only(handle->bdev)) {
+		bdev_release(handle);
+		return ERR_PTR(-EACCES);
+	}
+
+	return handle;
+}
+EXPORT_SYMBOL(bdev_open_by_path);
+
 void blkdev_put(struct block_device *bdev, void *holder)
 {
 	struct gendisk *disk = bdev->bd_disk;
@@ -903,6 +944,13 @@ void blkdev_put(struct block_device *bdev, void *holder)
 }
 EXPORT_SYMBOL(blkdev_put);
 
+void bdev_release(struct bdev_handle *handle)
+{
+	blkdev_put(handle->bdev, handle->holder);
+	kfree(handle);
+}
+EXPORT_SYMBOL(bdev_release);
+
 /**
  * lookup_bdev() - Look up a struct block_device by name.
  * @pathname: Name of the block device in the filesystem.
include/linux/blkdev.h +10 -0
index fa462d48ee96..6a942ec773b6 100644
@@ -1478,14 +1478,24 @@ extern const struct blk_holder_ops fs_holder_ops;
 #define sb_open_mode(flags) \
 	(BLK_OPEN_READ | (((flags) & SB_RDONLY) ? 0 : BLK_OPEN_WRITE))
 
+struct bdev_handle {
+	struct block_device *bdev;
+	void *holder;
+};
+
 struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 		const struct blk_holder_ops *hops);
 struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
 		void *holder, const struct blk_holder_ops *hops);
+struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
+		const struct blk_holder_ops *hops);
+struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
+		void *holder, const struct blk_holder_ops *hops);
 int bd_prepare_to_claim(struct block_device *bdev, void *holder,
 		const struct blk_holder_ops *hops);
 void bd_abort_claiming(struct block_device *bdev, void *holder);
 void blkdev_put(struct block_device *bdev, void *holder);
+void bdev_release(struct bdev_handle *handle);
 
 /* just for blk-cgroup, don't use elsewhere */
 struct block_device *blkdev_get_no_open(dev_t dev);
Signature · 1 line
2.35.3
maintainer Reviewed-by on v3 1/29
#2 lore raw

Re: [PATCH 01/29] block: Provide bdev_open_* functions

Quoted 9 lines · Jan Kara

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. This will eventually allow us to pass one more argument to blkdev_put() (renamed to bdev_release()) without too much hassle.

---

Looks good to me,

author reviewer
#3 lore raw

PATCH v3 1/29 block: Provide bdev_open_* functions

In reply to Jan Kara

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. This will eventually allow us to pass one more argument to blkdev_put() (renamed to bdev_release()) without too much hassle.

---
 block/bdev.c           | 48 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/blkdev.h | 10 +++++++++
 2 files changed, 58 insertions(+)
block/bdev.c +48 -0
index f3b13aa1b7d4..bdc7d739882b 100644
@@ -829,6 +829,25 @@ struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 }
 EXPORT_SYMBOL(blkdev_get_by_dev);
 
+struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
+				     const struct blk_holder_ops *hops)
+{
+	struct bdev_handle *handle = kmalloc(sizeof(*handle), GFP_KERNEL);
+	struct block_device *bdev;
+
+	if (!handle)
+		return ERR_PTR(-ENOMEM);
+	bdev = blkdev_get_by_dev(dev, mode, holder, hops);
+	if (IS_ERR(bdev)) {
+		kfree(handle);
+		return ERR_CAST(bdev);
+	}
+	handle->bdev = bdev;
+	handle->holder = holder;
+	return handle;
+}
+EXPORT_SYMBOL(bdev_open_by_dev);
+
 /**
  * blkdev_get_by_path - open a block device by name
  * @path: path to the block device to open
@@ -867,6 +886,28 @@ struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
 }
 EXPORT_SYMBOL(blkdev_get_by_path);
 
+struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
+		void *holder, const struct blk_holder_ops *hops)
+{
+	struct bdev_handle *handle;
+	dev_t dev;
+	int error;
+
+	error = lookup_bdev(path, &dev);
+	if (error)
+		return ERR_PTR(error);
+
+	handle = bdev_open_by_dev(dev, mode, holder, hops);
+	if (!IS_ERR(handle) && (mode & BLK_OPEN_WRITE) &&
+	    bdev_read_only(handle->bdev)) {
+		bdev_release(handle);
+		return ERR_PTR(-EACCES);
+	}
+
+	return handle;
+}
+EXPORT_SYMBOL(bdev_open_by_path);
+
 void blkdev_put(struct block_device *bdev, void *holder)
 {
 	struct gendisk *disk = bdev->bd_disk;
@@ -903,6 +944,13 @@ void blkdev_put(struct block_device *bdev, void *holder)
 }
 EXPORT_SYMBOL(blkdev_put);
 
+void bdev_release(struct bdev_handle *handle)
+{
+	blkdev_put(handle->bdev, handle->holder);
+	kfree(handle);
+}
+EXPORT_SYMBOL(bdev_release);
+
 /**
  * lookup_bdev() - Look up a struct block_device by name.
  * @pathname: Name of the block device in the filesystem.
include/linux/blkdev.h +10 -0
index eef450f25982..03d3adc3ff34 100644
@@ -1479,14 +1479,24 @@ extern const struct blk_holder_ops fs_holder_ops;
 #define sb_open_mode(flags) \
 	(BLK_OPEN_READ | (((flags) & SB_RDONLY) ? 0 : BLK_OPEN_WRITE))
 
+struct bdev_handle {
+	struct block_device *bdev;
+	void *holder;
+};
+
 struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 		const struct blk_holder_ops *hops);
 struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
 		void *holder, const struct blk_holder_ops *hops);
+struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
+		const struct blk_holder_ops *hops);
+struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
+		void *holder, const struct blk_holder_ops *hops);
 int bd_prepare_to_claim(struct block_device *bdev, void *holder,
 		const struct blk_holder_ops *hops);
 void bd_abort_claiming(struct block_device *bdev, void *holder);
 void blkdev_put(struct block_device *bdev, void *holder);
+void bdev_release(struct bdev_handle *handle);
 
 /* just for blk-cgroup, don't use elsewhere */
 struct block_device *blkdev_get_no_open(dev_t dev);
Signature · 1 line
2.35.3
author reviewer
#4 lore raw

PATCH v3 2/29 block: Use bdev_open_by_dev() in blkdev_open()

In reply to Jan Kara

Convert blkdev_open() to use bdev_open_by_dev(). To be able to propagate handle from blkdev_open() to blkdev_release() we need to stop using existence of file->private_data to determine exclusive block device opens. Use bdev_handle->mode for this purpose since file->f_flags isn't usable for this (O_EXCL is cleared from the flags during open).

---
 block/bdev.c           |  3 +++
 block/blk.h            |  1 -
 block/fops.c           | 34 ++++++++++++++++------------------
 block/ioctl.c          |  6 ++++--
 include/linux/blkdev.h |  1 +
 5 files changed, 24 insertions(+), 21 deletions(-)
block/bdev.c +3 -0
index fce150f9e66d..f1de1e65517b 100644
@@ -844,6 +844,9 @@ struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 	}
 	handle->bdev = bdev;
 	handle->holder = holder;
+	if (holder)
+		mode |= BLK_OPEN_EXCL;
+	handle->mode = mode;
 	return handle;
 }
 EXPORT_SYMBOL(bdev_open_by_dev);
block/blk.h +0 -1
index 608c5dcc516b..43b80dc78918 100644
@@ -464,7 +464,6 @@ extern struct device_attribute dev_attr_events_poll_msecs;
 
 extern struct attribute_group blk_trace_attr_group;
 
-blk_mode_t file_to_blk_mode(struct file *file);
 int truncate_bdev_range(struct block_device *bdev, blk_mode_t mode,
 		loff_t lstart, loff_t lend);
 long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
block/fops.c +16 -18
index a286bf3325c5..7261c27c7901 100644
@@ -470,7 +470,7 @@ static int blkdev_fsync(struct file *filp, loff_t start, loff_t end,
 	return error;
 }
 
-blk_mode_t file_to_blk_mode(struct file *file)
+static blk_mode_t file_to_blk_mode(struct file *file)
 {
 	blk_mode_t mode = 0;
 
@@ -478,7 +478,7 @@ blk_mode_t file_to_blk_mode(struct file *file)
 		mode |= BLK_OPEN_READ;
 	if (file->f_mode & FMODE_WRITE)
 		mode |= BLK_OPEN_WRITE;
-	if (file->private_data)
+	if (file->f_flags & O_EXCL)
 		mode |= BLK_OPEN_EXCL;
 	if (file->f_flags & O_NDELAY)
 		mode |= BLK_OPEN_NDELAY;
@@ -496,7 +496,8 @@ blk_mode_t file_to_blk_mode(struct file *file)
 
 static int blkdev_open(struct inode *inode, struct file *filp)
 {
-	struct block_device *bdev;
+	struct bdev_handle *handle;
+	blk_mode_t mode;
 
 	/*
 	 * Preserve backwards compatibility and allow large file access
@@ -507,29 +508,24 @@ static int blkdev_open(struct inode *inode, struct file *filp)
 	filp->f_flags |= O_LARGEFILE;
 	filp->f_mode |= FMODE_BUF_RASYNC;
 
-	/*
-	 * Use the file private data to store the holder for exclusive openes.
-	 * file_to_blk_mode relies on it being present to set BLK_OPEN_EXCL.
-	 */
-	if (filp->f_flags & O_EXCL)
-		filp->private_data = filp;
-
-	bdev = blkdev_get_by_dev(inode->i_rdev, file_to_blk_mode(filp),
-				 filp->private_data, NULL);
-	if (IS_ERR(bdev))
-		return PTR_ERR(bdev);
+	mode = file_to_blk_mode(filp);
+	handle = bdev_open_by_dev(inode->i_rdev, mode,
+			mode & BLK_OPEN_EXCL ? filp : NULL, NULL);
+	if (IS_ERR(handle))
+		return PTR_ERR(handle);
 
-	if (bdev_nowait(bdev))
+	if (bdev_nowait(handle->bdev))
 		filp->f_mode |= FMODE_NOWAIT;
 
-	filp->f_mapping = bdev->bd_inode->i_mapping;
+	filp->f_mapping = handle->bdev->bd_inode->i_mapping;
 	filp->f_wb_err = filemap_sample_wb_err(filp->f_mapping);
+	filp->private_data = handle;
 	return 0;
 }
 
 static int blkdev_release(struct inode *inode, struct file *filp)
 {
-	blkdev_put(I_BDEV(filp->f_mapping->host), filp->private_data);
+	bdev_release(filp->private_data);
 	return 0;
 }
 
@@ -630,6 +626,8 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start,
 {
 	struct inode *inode = bdev_file_inode(file);
 	struct block_device *bdev = I_BDEV(inode);
+	struct bdev_handle *bdev_handle = file->private_data;
+	blk_mode_t open_mode = bdev_handle->mode;
 	loff_t end = start + len - 1;
 	loff_t isize;
 	int error;
@@ -659,7 +657,7 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start,
 	filemap_invalidate_lock(inode->i_mapping);
 
 	/* Invalidate the page cache, including dirty pages. */
-	error = truncate_bdev_range(bdev, file_to_blk_mode(file), start, end);
+	error = truncate_bdev_range(bdev, open_mode, start, end);
 	if (error)
 		goto fail;
 
block/ioctl.c +4 -2
index 648670ddb164..54c1e2f71031 100644
@@ -582,7 +582,8 @@ long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
 	struct block_device *bdev = I_BDEV(file->f_mapping->host);
 	void __user *argp = (void __user *)arg;
-	blk_mode_t mode = file_to_blk_mode(file);
+	struct bdev_handle *bdev_handle = file->private_data;
+	blk_mode_t mode = bdev_handle->mode;
 	int ret;
 
 	switch (cmd) {
@@ -643,7 +644,8 @@ long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 	void __user *argp = compat_ptr(arg);
 	struct block_device *bdev = I_BDEV(file->f_mapping->host);
 	struct gendisk *disk = bdev->bd_disk;
-	blk_mode_t mode = file_to_blk_mode(file);
+	struct bdev_handle *bdev_handle = file->private_data;
+	blk_mode_t mode = bdev_handle->mode;
 
 	switch (cmd) {
 	/* These need separate implementations for the data structure */
include/linux/blkdev.h +1 -0
index 6a942ec773b6..ae741dec184b 100644
@@ -1481,6 +1481,7 @@ extern const struct blk_holder_ops fs_holder_ops;
 struct bdev_handle {
 	struct block_device *bdev;
 	void *holder;
+	blk_mode_t mode;
 };
 
 struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
Signature · 1 line
2.35.3
maintainer
#5 lore raw

Re: [PATCH 02/29] block: Use bdev_open_by_dev() in blkdev_open()

Quoted 5 lines · Jan Kara
block/ioctl.c +0 -0
index 648670ddb164..54c1e2f71031 100644

block/ioctl.c long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) · line 589

Quoted 9 lines
block/ioctl.c +2 -1
@@ -582,7 +582,8 @@ long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
 	struct block_device *bdev = I_BDEV(file->f_mapping->host);
 	void __user *argp = (void __user *)arg;
-	blk_mode_t mode = file_to_blk_mode(file);
+	struct bdev_handle *bdev_handle = file->private_data;
+	blk_mode_t mode = bdev_handle->mode;
 	int ret;
 
 	switch (cmd) {

Still the same bug as in v2 - you are missing the effects of
fcntl(2) setting/clearing O_NDELAY and sd_ioctl() is sensitive to that.

author reviewer Acked-by Christoph Hellwig
#6 lore raw

Re: [PATCH 02/29] block: Use bdev_open_by_dev() in blkdev_open()

Quoted 14 lines · Al Viro
Quoted 13 lines · Jan Kara
block/ioctl.c +2 -1
index 648670ddb164..54c1e2f71031 100644
@@ -582,7 +582,8 @@ long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 {
 	struct block_device *bdev = I_BDEV(file->f_mapping->host);
 	void __user *argp = (void __user *)arg;
-	blk_mode_t mode = file_to_blk_mode(file);
+	struct bdev_handle *bdev_handle = file->private_data;
+	blk_mode_t mode = bdev_handle->mode;
 	int ret;
 
Quoted 4 lines

block/ioctl.c long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) · line 589

Quoted 1 line
block/ioctl.c +0 -0
 	switch (cmd) {

Still the same bug as in v2 - you are missing the effects of
fcntl(2) setting/clearing O_NDELAY and sd_ioctl() is sensitive to that.

Argh, indeed you are correct. I forgot about fcntl(2) modifying file->f_flags. Attached is a version of the patch that I'm currently testing.

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR

From 46df2340545e872d34cf824e0d33b9b00b91e32f Mon Sep 17 00:00:00 2001
From: Jan Kara <[email protected]>
Date: Wed, 21 Jun 2023 13:16:32 +0200
Subject: [PATCH] block: Use bdev_open_by_dev() in blkdev_open()

Convert blkdev_open() to use bdev_open_by_dev(). To be able to propagate handle from blkdev_open() to blkdev_release() we need to stop using existence of file->private_data to determine exclusive block device opens. Use bdev_handle->mode for this purpose since file->f_flags isn't usable for this (O_EXCL is cleared from the flags during open).

---
 block/bdev.c           |  3 +++
 block/fops.c           | 35 +++++++++++++++++++----------------
 include/linux/blkdev.h |  1 +
 3 files changed, 23 insertions(+), 16 deletions(-)
block/bdev.c +3 -0
index fce150f9e66d..f1de1e65517b 100644
@@ -844,6 +844,9 @@ struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 	}
 	handle->bdev = bdev;
 	handle->holder = holder;
+	if (holder)
+		mode |= BLK_OPEN_EXCL;
+	handle->mode = mode;
 	return handle;
 }
 EXPORT_SYMBOL(bdev_open_by_dev);
block/fops.c +19 -16
index a286bf3325c5..5b0e6899c5ad 100644
@@ -473,12 +473,19 @@ static int blkdev_fsync(struct file *filp, loff_t start, loff_t end,
 blk_mode_t file_to_blk_mode(struct file *file)
 {
 	blk_mode_t mode = 0;
+	struct bdev_handle *handle = file->private_data;
 
 	if (file->f_mode & FMODE_READ)
 		mode |= BLK_OPEN_READ;
 	if (file->f_mode & FMODE_WRITE)
 		mode |= BLK_OPEN_WRITE;
-	if (file->private_data)
+	/*
+	 * do_dentry_open() clears O_EXCL from f_flags, use handle->mode to
+	 * determine whether the open was exclusive for already open files.
+	 */
+	if (handle)
+		mode |= handle->mode & BLK_OPEN_EXCL;
+	else if (file->f_flags & O_EXCL)
 		mode |= BLK_OPEN_EXCL;
 	if (file->f_flags & O_NDELAY)
 		mode |= BLK_OPEN_NDELAY;
@@ -496,7 +503,8 @@ blk_mode_t file_to_blk_mode(struct file *file)
 
 static int blkdev_open(struct inode *inode, struct file *filp)
 {
-	struct block_device *bdev;
+	struct bdev_handle *handle;
+	blk_mode_t mode;
 
 	/*
 	 * Preserve backwards compatibility and allow large file access
@@ -507,29 +515,24 @@ static int blkdev_open(struct inode *inode, struct file *filp)
 	filp->f_flags |= O_LARGEFILE;
 	filp->f_mode |= FMODE_BUF_RASYNC;
 
-	/*
-	 * Use the file private data to store the holder for exclusive openes.
-	 * file_to_blk_mode relies on it being present to set BLK_OPEN_EXCL.
-	 */
-	if (filp->f_flags & O_EXCL)
-		filp->private_data = filp;
-
-	bdev = blkdev_get_by_dev(inode->i_rdev, file_to_blk_mode(filp),
-				 filp->private_data, NULL);
-	if (IS_ERR(bdev))
-		return PTR_ERR(bdev);
+	mode = file_to_blk_mode(filp);
+	handle = bdev_open_by_dev(inode->i_rdev, mode,
+			mode & BLK_OPEN_EXCL ? filp : NULL, NULL);
+	if (IS_ERR(handle))
+		return PTR_ERR(handle);
 
-	if (bdev_nowait(bdev))
+	if (bdev_nowait(handle->bdev))
 		filp->f_mode |= FMODE_NOWAIT;
 
-	filp->f_mapping = bdev->bd_inode->i_mapping;
+	filp->f_mapping = handle->bdev->bd_inode->i_mapping;
 	filp->f_wb_err = filemap_sample_wb_err(filp->f_mapping);
+	filp->private_data = handle;
 	return 0;
 }
 
 static int blkdev_release(struct inode *inode, struct file *filp)
 {
-	blkdev_put(I_BDEV(filp->f_mapping->host), filp->private_data);
+	bdev_release(filp->private_data);
 	return 0;
 }
 
include/linux/blkdev.h +1 -0
index 6a942ec773b6..ae741dec184b 100644
@@ -1481,6 +1481,7 @@ extern const struct blk_holder_ops fs_holder_ops;
 struct bdev_handle {
 	struct block_device *bdev;
 	void *holder;
+	blk_mode_t mode;
 };
 
 struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
Signature · 1 line
2.35.3
maintainer Reviewed-by
#7 lore raw

Re: [PATCH 02/29] block: Use bdev_open_by_dev() in blkdev_open()

Quoted 2 lines

file->f_flags. Attached is a version of the patch that I'm currently testing.

Appended patch looks good to me,

The patch also has another fix for O_EXCL. In earlier versions of this patch series it relied of f_flags. Thanks for that comment you added in there about this now. This really helps given that O_EXCL has special meaning for block devices. Ideally we'd have kernel doc for file_to_blk_mode().

author reviewer
#8 lore raw

Re: [PATCH 02/29] block: Use bdev_open_by_dev() in blkdev_open()

Quoted 11 lines · Christian Brauner
Quoted 2 lines

file->f_flags. Attached is a version of the patch that I'm currently testing.

Appended patch looks good to me,

The patch also has another fix for O_EXCL. In earlier versions of this patch series it relied of f_flags. Thanks for that comment you added in there about this now. This really helps given that O_EXCL has special meaning for block devices. Ideally we'd have kernel doc for file_to_blk_mode().

Thanks for review! I've added the kerneldoc comment:

/**
* file_to_blk_mode - get block open flags from file flags
* @file: file whose open flags should be converted
*
* Look at file open flags and generate corresponding block open flags from
* them. The function works both for file just being open (e.g. during ->open
* callback) and for file that is already open. This is actually non-trivial
* (see comment in the function).
*/

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
#9 lore raw

Re: [PATCH 02/29] block: Use bdev_open_by_dev() in blkdev_open()

Quoted 15 lines · Jan Kara
Quoted 11 lines · Christian Brauner
Quoted 2 lines

file->f_flags. Attached is a version of the patch that I'm currently testing.

Appended patch looks good to me,

The patch also has another fix for O_EXCL. In earlier versions of this patch series it relied of f_flags. Thanks for that comment you added in there about this now. This really helps given that O_EXCL has special meaning for block devices. Ideally we'd have kernel doc for file_to_blk_mode().

Thanks for review! I've added the kerneldoc comment:

Quoted 9 lines

/**
* file_to_blk_mode - get block open flags from file flags
* @file: file whose open flags should be converted
*
* Look at file open flags and generate corresponding block open flags from
* them. The function works both for file just being open (e.g. during ->open
* callback) and for file that is already open. This is actually non-trivial
* (see comment in the function).
*/

Perfect, thanks!

author reviewer
#10 lore raw

PATCH v3 2/29 block: Use bdev_open_by_dev() in blkdev_open()

In reply to Jan Kara

Convert blkdev_open() to use bdev_open_by_dev(). To be able to propagate handle from blkdev_open() to blkdev_release() we need to stop using existence of file->private_data to determine exclusive block device opens. Use bdev_handle->mode for this purpose since file->f_flags isn't usable for this (O_EXCL is cleared from the flags during open).

---
 block/bdev.c           |  3 +++
 block/fops.c           | 44 +++++++++++++++++++++++++++---------------
 include/linux/blkdev.h |  1 +
 3 files changed, 32 insertions(+), 16 deletions(-)
block/bdev.c +3 -0
index bdc7d739882b..4628dcb1da8a 100644
@@ -844,6 +844,9 @@ struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 	}
 	handle->bdev = bdev;
 	handle->holder = holder;
+	if (holder)
+		mode |= BLK_OPEN_EXCL;
+	handle->mode = mode;
 	return handle;
 }
 EXPORT_SYMBOL(bdev_open_by_dev);
block/fops.c +28 -16

79 lines not on this page · raw message

include/linux/blkdev.h +1 -0
index 03d3adc3ff34..51fa7ffdee83 100644
@@ -1482,6 +1482,7 @@ extern const struct blk_holder_ops fs_holder_ops;
 struct bdev_handle {
 	struct block_device *bdev;
 	void *holder;
+	blk_mode_t mode;
 };
 
 struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
Signature · 1 line
2.35.3
author reviewer
#11 lore raw

PATCH v3 3/29 block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset()

In reply to Jan Kara

Convert disk_scan_partitions() and blkdev_bszset() to use bdev_open_by_dev().

---
 block/genhd.c | 12 ++++++------
 block/ioctl.c |  6 ++++--
 2 files changed, 10 insertions(+), 8 deletions(-)
block/genhd.c +6 -6
index cc32a0c704eb..4a16a424f57d 100644
@@ -342,7 +342,7 @@ EXPORT_SYMBOL_GPL(disk_uevent);
 
 int disk_scan_partitions(struct gendisk *disk, blk_mode_t mode)
 {
-	struct block_device *bdev;
+	struct bdev_handle *handle;
 	int ret = 0;
 
 	if (disk->flags & (GENHD_FL_NO_PART | GENHD_FL_HIDDEN))
@@ -366,12 +366,12 @@ int disk_scan_partitions(struct gendisk *disk, blk_mode_t mode)
 	}
 
 	set_bit(GD_NEED_PART_SCAN, &disk->state);
-	bdev = blkdev_get_by_dev(disk_devt(disk), mode & ~BLK_OPEN_EXCL, NULL,
-				 NULL);
-	if (IS_ERR(bdev))
-		ret =  PTR_ERR(bdev);
+	handle = bdev_open_by_dev(disk_devt(disk), mode & ~BLK_OPEN_EXCL, NULL,
+				  NULL);
+	if (IS_ERR(handle))
+		ret = PTR_ERR(handle);
 	else
-		blkdev_put(bdev, NULL);
+		bdev_release(handle);
 
 	/*
 	 * If blkdev_get_by_dev() failed early, GD_NEED_PART_SCAN is still set,
block/ioctl.c +4 -2
index 54c1e2f71031..2e46dc38424c 100644
@@ -465,6 +465,7 @@ static int blkdev_bszset(struct block_device *bdev, blk_mode_t mode,
 		int __user *argp)
 {
 	int ret, n;
+	struct bdev_handle *handle;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -476,10 +477,11 @@ static int blkdev_bszset(struct block_device *bdev, blk_mode_t mode,
 	if (mode & BLK_OPEN_EXCL)
 		return set_blocksize(bdev, n);
 
-	if (IS_ERR(blkdev_get_by_dev(bdev->bd_dev, mode, &bdev, NULL)))
+	handle = bdev_open_by_dev(bdev->bd_dev, mode, &bdev, NULL);
+	if (IS_ERR(handle))
 		return -EBUSY;
 	ret = set_blocksize(bdev, n);
-	blkdev_put(bdev, &bdev);
+	bdev_release(handle);
 
 	return ret;
 }
Signature · 1 line
2.35.3
author reviewer
#13 lore raw

PATCH v3 3/29 block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset()

In reply to Jan Kara

Convert disk_scan_partitions() and blkdev_bszset() to use bdev_open_by_dev().

---
 block/genhd.c | 12 ++++++------
 block/ioctl.c |  6 ++++--
 2 files changed, 10 insertions(+), 8 deletions(-)
block/genhd.c +6 -6
index cc32a0c704eb..4a16a424f57d 100644
@@ -342,7 +342,7 @@ EXPORT_SYMBOL_GPL(disk_uevent);
 
 int disk_scan_partitions(struct gendisk *disk, blk_mode_t mode)
 {
-	struct block_device *bdev;
+	struct bdev_handle *handle;
 	int ret = 0;
 
 	if (disk->flags & (GENHD_FL_NO_PART | GENHD_FL_HIDDEN))
@@ -366,12 +366,12 @@ int disk_scan_partitions(struct gendisk *disk, blk_mode_t mode)
 	}
 
 	set_bit(GD_NEED_PART_SCAN, &disk->state);
-	bdev = blkdev_get_by_dev(disk_devt(disk), mode & ~BLK_OPEN_EXCL, NULL,
-				 NULL);
-	if (IS_ERR(bdev))
-		ret =  PTR_ERR(bdev);
+	handle = bdev_open_by_dev(disk_devt(disk), mode & ~BLK_OPEN_EXCL, NULL,
+				  NULL);
+	if (IS_ERR(handle))
+		ret = PTR_ERR(handle);
 	else
-		blkdev_put(bdev, NULL);
+		bdev_release(handle);
 
 	/*
 	 * If blkdev_get_by_dev() failed early, GD_NEED_PART_SCAN is still set,
block/ioctl.c +4 -2
index d5f5cd61efd7..5d356c964352 100644
@@ -467,6 +467,7 @@ static int blkdev_bszset(struct block_device *bdev, blk_mode_t mode,
 		int __user *argp)
 {
 	int ret, n;
+	struct bdev_handle *handle;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -478,10 +479,11 @@ static int blkdev_bszset(struct block_device *bdev, blk_mode_t mode,
 	if (mode & BLK_OPEN_EXCL)
 		return set_blocksize(bdev, n);
 
-	if (IS_ERR(blkdev_get_by_dev(bdev->bd_dev, mode, &bdev, NULL)))
+	handle = bdev_open_by_dev(bdev->bd_dev, mode, &bdev, NULL);
+	if (IS_ERR(handle))
 		return -EBUSY;
 	ret = set_blocksize(bdev, n);
-	blkdev_put(bdev, &bdev);
+	bdev_release(handle);
 
 	return ret;
 }
Signature · 1 line
2.35.3
author reviewer
#14 lore raw

PATCH v3 4/29 drdb: Convert to use bdev_open_by_path()

In reply to Jan Kara

Convert drdb to use bdev_open_by_path().

---
 drivers/block/drbd/drbd_int.h |  2 ++
 drivers/block/drbd/drbd_nl.c  | 65 +++++++++++++++++------------------
 2 files changed, 34 insertions(+), 33 deletions(-)
drivers/block/drbd/drbd_int.h +2 -0
index a30a5ed811be..f017e917612b 100644
@@ -524,7 +524,9 @@ struct drbd_md {
 
 struct drbd_backing_dev {
 	struct block_device *backing_bdev;
+	struct bdev_handle *backing_bdev_handle;
 	struct block_device *md_bdev;
+	struct bdev_handle *md_bdev_handle;
 	struct drbd_md md;
 	struct disk_conf *disk_conf; /* RCU, for updates: resource->conf_update */
 	sector_t known_size; /* last known size of that backing device */
drivers/block/drbd/drbd_nl.c +32 -33

131 lines not on this page · raw message

Signature · 1 line
2.35.3
author reviewer
#16 lore raw

PATCH v3 4/29 drdb: Convert to use bdev_open_by_path()

In reply to Jan Kara

Convert drdb to use bdev_open_by_path().

---
 drivers/block/drbd/drbd_int.h |  2 ++
 drivers/block/drbd/drbd_nl.c  | 65 +++++++++++++++++------------------
 2 files changed, 34 insertions(+), 33 deletions(-)
drivers/block/drbd/drbd_int.h +2 -0
index a30a5ed811be..f017e917612b 100644
@@ -524,7 +524,9 @@ struct drbd_md {
 
 struct drbd_backing_dev {
 	struct block_device *backing_bdev;
+	struct bdev_handle *backing_bdev_handle;
 	struct block_device *md_bdev;
+	struct bdev_handle *md_bdev_handle;
 	struct drbd_md md;
 	struct disk_conf *disk_conf; /* RCU, for updates: resource->conf_update */
 	sector_t known_size; /* last known size of that backing device */
drivers/block/drbd/drbd_nl.c +32 -33

131 lines not on this page · raw message

Signature · 1 line
2.35.3
author reviewer
#17 lore raw

PATCH v3 5/29 pktcdvd: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert pktcdvd to use bdev_open_by_dev().

---
 drivers/block/pktcdvd.c | 76 ++++++++++++++++++++++-------------------
 include/linux/pktcdvd.h |  4 ++-
 2 files changed, 44 insertions(+), 36 deletions(-)
drivers/block/pktcdvd.c +41 -35
index a1428538bda5..d56d972aadb3 100644
@@ -340,8 +340,8 @@ static ssize_t device_map_show(const struct class *c, const struct class_attribu
 		n += sysfs_emit_at(data, n, "%s %u:%u %u:%u\n",
 			pd->disk->disk_name,
 			MAJOR(pd->pkt_dev), MINOR(pd->pkt_dev),
-			MAJOR(pd->bdev->bd_dev),
-			MINOR(pd->bdev->bd_dev));
+			MAJOR(pd->bdev_handle->bdev->bd_dev),
+			MINOR(pd->bdev_handle->bdev->bd_dev));
 	}
 	mutex_unlock(&ctl_mutex);
 	return n;
@@ -437,7 +437,8 @@ static int pkt_seq_show(struct seq_file *m, void *p)
 	char *msg;
 	int states[PACKET_NUM_STATES];
 
-	seq_printf(m, "Writer %s mapped to %pg:\n", pd->disk->disk_name, pd->bdev);
+	seq_printf(m, "Writer %s mapped to %pg:\n", pd->disk->disk_name,
+		   pd->bdev_handle->bdev);
 
 	seq_printf(m, "\nSettings:\n");
 	seq_printf(m, "\tpacket size:\t\t%dkB\n", pd->settings.size / 2);
@@ -714,7 +715,7 @@ static void pkt_rbtree_insert(struct pktcdvd_device *pd, struct pkt_rb_node *nod
  */
 static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *cgc)
 {
-	struct request_queue *q = bdev_get_queue(pd->bdev);
+	struct request_queue *q = bdev_get_queue(pd->bdev_handle->bdev);
 	struct scsi_cmnd *scmd;
 	struct request *rq;
 	int ret = 0;
@@ -1047,7 +1048,8 @@ static void pkt_gather_data(struct pktcdvd_device *pd, struct packet_data *pkt)
 			continue;
 
 		bio = pkt->r_bios[f];
-		bio_init(bio, pd->bdev, bio->bi_inline_vecs, 1, REQ_OP_READ);
+		bio_init(bio, pd->bdev_handle->bdev, bio->bi_inline_vecs, 1,
+			 REQ_OP_READ);
 		bio->bi_iter.bi_sector = pkt->sector + f * (CD_FRAMESIZE >> 9);
 		bio->bi_end_io = pkt_end_io_read;
 		bio->bi_private = pkt;
@@ -1262,8 +1264,8 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 	struct device *ddev = disk_to_dev(pd->disk);
 	int f;
 
-	bio_init(pkt->w_bio, pd->bdev, pkt->w_bio->bi_inline_vecs, pkt->frames,
-		 REQ_OP_WRITE);
+	bio_init(pkt->w_bio, pd->bdev_handle->bdev, pkt->w_bio->bi_inline_vecs,
+		 pkt->frames, REQ_OP_WRITE);
 	pkt->w_bio->bi_iter.bi_sector = pkt->sector;
 	pkt->w_bio->bi_end_io = pkt_end_io_packet_write;
 	pkt->w_bio->bi_private = pkt;
@@ -2160,18 +2162,20 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write)
 	int ret;
 	long lba;
 	struct request_queue *q;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 
 	/*
 	 * We need to re-open the cdrom device without O_NONBLOCK to be able
 	 * to read/write from/to it. It is already opened in O_NONBLOCK mode
 	 * so open should not fail.
 	 */
-	bdev = blkdev_get_by_dev(pd->bdev->bd_dev, BLK_OPEN_READ, pd, NULL);
-	if (IS_ERR(bdev)) {
-		ret = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_dev(pd->bdev_handle->bdev->bd_dev,
+				       BLK_OPEN_READ, pd, NULL);
+	if (IS_ERR(bdev_handle)) {
+		ret = PTR_ERR(bdev_handle);
 		goto out;
 	}
+	pd->open_bdev_handle = bdev_handle;
 
 	ret = pkt_get_last_written(pd, &lba);
 	if (ret) {
@@ -2180,9 +2184,9 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write)
 	}
 
 	set_capacity(pd->disk, lba << 2);
-	set_capacity_and_notify(pd->bdev->bd_disk, lba << 2);
+	set_capacity_and_notify(pd->bdev_handle->bdev->bd_disk, lba << 2);
 
-	q = bdev_get_queue(pd->bdev);
+	q = bdev_get_queue(pd->bdev_handle->bdev);
 	if (write) {
 		ret = pkt_open_write(pd);
 		if (ret)
@@ -2214,7 +2218,7 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write)
 	return 0;
 
 out_putdev:
-	blkdev_put(bdev, pd);
+	bdev_release(bdev_handle);
 out:
 	return ret;
 }
@@ -2233,7 +2237,8 @@ static void pkt_release_dev(struct pktcdvd_device *pd, int flush)
 	pkt_lock_door(pd, 0);
 
 	pkt_set_speed(pd, MAX_SPEED, MAX_SPEED);
-	blkdev_put(pd->bdev, pd);
+	bdev_release(pd->open_bdev_handle);
+	pd->open_bdev_handle = NULL;
 
 	pkt_shrink_pktlist(pd);
 }
@@ -2321,8 +2326,8 @@ static void pkt_end_io_read_cloned(struct bio *bio)
 
 static void pkt_make_request_read(struct pktcdvd_device *pd, struct bio *bio)
 {
-	struct bio *cloned_bio =
-		bio_alloc_clone(pd->bdev, bio, GFP_NOIO, &pkt_bio_set);
+	struct bio *cloned_bio = bio_alloc_clone(pd->bdev_handle->bdev, bio,
+		GFP_NOIO, &pkt_bio_set);
 	struct packet_stacked_data *psd = mempool_alloc(&psd_pool, GFP_NOIO);
 
 	psd->pd = pd;
@@ -2492,7 +2497,7 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 {
 	struct device *ddev = disk_to_dev(pd->disk);
 	int i;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct scsi_device *sdev;
 
 	if (pd->pkt_dev == dev) {
@@ -2503,8 +2508,9 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 		struct pktcdvd_device *pd2 = pkt_devs[i];
 		if (!pd2)
 			continue;
-		if (pd2->bdev->bd_dev == dev) {
-			dev_err(ddev, "%pg already setup\n", pd2->bdev);
+		if (pd2->bdev_handle->bdev->bd_dev == dev) {
+			dev_err(ddev, "%pg already setup\n",
+				pd2->bdev_handle->bdev);
 			return -EBUSY;
 		}
 		if (pd2->pkt_dev == dev) {
@@ -2513,13 +2519,13 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 		}
 	}
 
-	bdev = blkdev_get_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_NDELAY, NULL,
-				 NULL);
-	if (IS_ERR(bdev))
-		return PTR_ERR(bdev);
-	sdev = scsi_device_from_queue(bdev->bd_disk->queue);
+	bdev_handle = bdev_open_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_NDELAY,
+				       NULL, NULL);
+	if (IS_ERR(bdev_handle))
+		return PTR_ERR(bdev_handle);
+	sdev = scsi_device_from_queue(bdev_handle->bdev->bd_disk->queue);
 	if (!sdev) {
-		blkdev_put(bdev, NULL);
+		bdev_release(bdev_handle);
 		return -EINVAL;
 	}
 	put_device(&sdev->sdev_gendev);
@@ -2527,8 +2533,8 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 	/* This is safe, since we have a reference from open(). */
 	__module_get(THIS_MODULE);
 
-	pd->bdev = bdev;
-	set_blocksize(bdev, CD_FRAMESIZE);
+	pd->bdev_handle = bdev_handle;
+	set_blocksize(bdev_handle->bdev, CD_FRAMESIZE);
 
 	pkt_init_queue(pd);
 
@@ -2540,11 +2546,11 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 	}
 
 	proc_create_single_data(pd->disk->disk_name, 0, pkt_proc, pkt_seq_show, pd);
-	dev_notice(ddev, "writer mapped to %pg\n", bdev);
+	dev_notice(ddev, "writer mapped to %pg\n", bdev_handle->bdev);
 	return 0;
 
 out_mem:
-	blkdev_put(bdev, NULL);
+	bdev_release(bdev_handle);
 	/* This is safe: open() is still holding a reference. */
 	module_put(THIS_MODULE);
 	return -ENOMEM;
@@ -2599,9 +2605,9 @@ static unsigned int pkt_check_events(struct gendisk *disk,
 
 	if (!pd)
 		return 0;
-	if (!pd->bdev)
+	if (!pd->bdev_handle)
 		return 0;
-	attached_disk = pd->bdev->bd_disk;
+	attached_disk = pd->bdev_handle->bdev->bd_disk;
 	if (!attached_disk || !attached_disk->fops->check_events)
 		return 0;
 	return attached_disk->fops->check_events(attached_disk, clearing);
@@ -2686,7 +2692,7 @@ static int pkt_setup_dev(dev_t dev, dev_t* pkt_dev)
 		goto out_mem2;
 
 	/* inherit events of the host device */
-	disk->events = pd->bdev->bd_disk->events;
+	disk->events = pd->bdev_handle->bdev->bd_disk->events;
 
 	ret = add_disk(disk);
 	if (ret)
@@ -2751,7 +2757,7 @@ static int pkt_remove_dev(dev_t pkt_dev)
 	pkt_debugfs_dev_remove(pd);
 	pkt_sysfs_dev_remove(pd);
 
-	blkdev_put(pd->bdev, NULL);
+	bdev_release(pd->bdev_handle);
 
 	remove_proc_entry(pd->disk->disk_name, pkt_proc);
 	dev_notice(ddev, "writer unmapped\n");
@@ -2778,7 +2784,7 @@ static void pkt_get_status(struct pkt_ctrl_command *ctrl_cmd)
 
 	pd = pkt_find_dev_from_minor(ctrl_cmd->dev_index);
 	if (pd) {
-		ctrl_cmd->dev = new_encode_dev(pd->bdev->bd_dev);
+		ctrl_cmd->dev = new_encode_dev(pd->bdev_handle->bdev->bd_dev);
 		ctrl_cmd->pkt_dev = new_encode_dev(pd->pkt_dev);
 	} else {
 		ctrl_cmd->dev = 0;
include/linux/pktcdvd.h +3 -1
index 80cb00db42a4..79594aeb160d 100644
@@ -154,7 +154,9 @@ struct packet_stacked_data
 
 struct pktcdvd_device
 {
-	struct block_device	*bdev;		/* dev attached */
+	struct bdev_handle	*bdev_handle;	/* dev attached */
+	/* handle acquired for bdev during pkt_open_dev() */
+	struct bdev_handle	*open_bdev_handle;
 	dev_t			pkt_dev;	/* our dev */
 	struct packet_settings	settings;
 	struct packet_stats	stats;
Signature · 1 line
2.35.3
author reviewer
#19 lore raw

PATCH v3 5/29 pktcdvd: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert pktcdvd to use bdev_open_by_dev().

---
 drivers/block/pktcdvd.c | 76 ++++++++++++++++++++++-------------------
 include/linux/pktcdvd.h |  4 ++-
 2 files changed, 44 insertions(+), 36 deletions(-)
drivers/block/pktcdvd.c +41 -35
index a1428538bda5..d56d972aadb3 100644
@@ -340,8 +340,8 @@ static ssize_t device_map_show(const struct class *c, const struct class_attribu
 		n += sysfs_emit_at(data, n, "%s %u:%u %u:%u\n",
 			pd->disk->disk_name,
 			MAJOR(pd->pkt_dev), MINOR(pd->pkt_dev),
-			MAJOR(pd->bdev->bd_dev),
-			MINOR(pd->bdev->bd_dev));
+			MAJOR(pd->bdev_handle->bdev->bd_dev),
+			MINOR(pd->bdev_handle->bdev->bd_dev));
 	}
 	mutex_unlock(&ctl_mutex);
 	return n;
@@ -437,7 +437,8 @@ static int pkt_seq_show(struct seq_file *m, void *p)
 	char *msg;
 	int states[PACKET_NUM_STATES];
 
-	seq_printf(m, "Writer %s mapped to %pg:\n", pd->disk->disk_name, pd->bdev);
+	seq_printf(m, "Writer %s mapped to %pg:\n", pd->disk->disk_name,
+		   pd->bdev_handle->bdev);
 
 	seq_printf(m, "\nSettings:\n");
 	seq_printf(m, "\tpacket size:\t\t%dkB\n", pd->settings.size / 2);
@@ -714,7 +715,7 @@ static void pkt_rbtree_insert(struct pktcdvd_device *pd, struct pkt_rb_node *nod
  */
 static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *cgc)
 {
-	struct request_queue *q = bdev_get_queue(pd->bdev);
+	struct request_queue *q = bdev_get_queue(pd->bdev_handle->bdev);
 	struct scsi_cmnd *scmd;
 	struct request *rq;
 	int ret = 0;
@@ -1047,7 +1048,8 @@ static void pkt_gather_data(struct pktcdvd_device *pd, struct packet_data *pkt)
 			continue;
 
 		bio = pkt->r_bios[f];
-		bio_init(bio, pd->bdev, bio->bi_inline_vecs, 1, REQ_OP_READ);
+		bio_init(bio, pd->bdev_handle->bdev, bio->bi_inline_vecs, 1,
+			 REQ_OP_READ);
 		bio->bi_iter.bi_sector = pkt->sector + f * (CD_FRAMESIZE >> 9);
 		bio->bi_end_io = pkt_end_io_read;
 		bio->bi_private = pkt;
@@ -1262,8 +1264,8 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 	struct device *ddev = disk_to_dev(pd->disk);
 	int f;
 
-	bio_init(pkt->w_bio, pd->bdev, pkt->w_bio->bi_inline_vecs, pkt->frames,
-		 REQ_OP_WRITE);
+	bio_init(pkt->w_bio, pd->bdev_handle->bdev, pkt->w_bio->bi_inline_vecs,
+		 pkt->frames, REQ_OP_WRITE);
 	pkt->w_bio->bi_iter.bi_sector = pkt->sector;
 	pkt->w_bio->bi_end_io = pkt_end_io_packet_write;
 	pkt->w_bio->bi_private = pkt;
@@ -2160,18 +2162,20 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write)
 	int ret;
 	long lba;
 	struct request_queue *q;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 
 	/*
 	 * We need to re-open the cdrom device without O_NONBLOCK to be able
 	 * to read/write from/to it. It is already opened in O_NONBLOCK mode
 	 * so open should not fail.
 	 */
-	bdev = blkdev_get_by_dev(pd->bdev->bd_dev, BLK_OPEN_READ, pd, NULL);
-	if (IS_ERR(bdev)) {
-		ret = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_dev(pd->bdev_handle->bdev->bd_dev,
+				       BLK_OPEN_READ, pd, NULL);
+	if (IS_ERR(bdev_handle)) {
+		ret = PTR_ERR(bdev_handle);
 		goto out;
 	}
+	pd->open_bdev_handle = bdev_handle;
 
 	ret = pkt_get_last_written(pd, &lba);
 	if (ret) {
@@ -2180,9 +2184,9 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write)
 	}
 
 	set_capacity(pd->disk, lba << 2);
-	set_capacity_and_notify(pd->bdev->bd_disk, lba << 2);
+	set_capacity_and_notify(pd->bdev_handle->bdev->bd_disk, lba << 2);
 
-	q = bdev_get_queue(pd->bdev);
+	q = bdev_get_queue(pd->bdev_handle->bdev);
 	if (write) {
 		ret = pkt_open_write(pd);
 		if (ret)
@@ -2214,7 +2218,7 @@ static int pkt_open_dev(struct pktcdvd_device *pd, bool write)
 	return 0;
 
 out_putdev:
-	blkdev_put(bdev, pd);
+	bdev_release(bdev_handle);
 out:
 	return ret;
 }
@@ -2233,7 +2237,8 @@ static void pkt_release_dev(struct pktcdvd_device *pd, int flush)
 	pkt_lock_door(pd, 0);
 
 	pkt_set_speed(pd, MAX_SPEED, MAX_SPEED);
-	blkdev_put(pd->bdev, pd);
+	bdev_release(pd->open_bdev_handle);
+	pd->open_bdev_handle = NULL;
 
 	pkt_shrink_pktlist(pd);
 }
@@ -2321,8 +2326,8 @@ static void pkt_end_io_read_cloned(struct bio *bio)
 
 static void pkt_make_request_read(struct pktcdvd_device *pd, struct bio *bio)
 {
-	struct bio *cloned_bio =
-		bio_alloc_clone(pd->bdev, bio, GFP_NOIO, &pkt_bio_set);
+	struct bio *cloned_bio = bio_alloc_clone(pd->bdev_handle->bdev, bio,
+		GFP_NOIO, &pkt_bio_set);
 	struct packet_stacked_data *psd = mempool_alloc(&psd_pool, GFP_NOIO);
 
 	psd->pd = pd;
@@ -2492,7 +2497,7 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 {
 	struct device *ddev = disk_to_dev(pd->disk);
 	int i;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct scsi_device *sdev;
 
 	if (pd->pkt_dev == dev) {
@@ -2503,8 +2508,9 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 		struct pktcdvd_device *pd2 = pkt_devs[i];
 		if (!pd2)
 			continue;
-		if (pd2->bdev->bd_dev == dev) {
-			dev_err(ddev, "%pg already setup\n", pd2->bdev);
+		if (pd2->bdev_handle->bdev->bd_dev == dev) {
+			dev_err(ddev, "%pg already setup\n",
+				pd2->bdev_handle->bdev);
 			return -EBUSY;
 		}
 		if (pd2->pkt_dev == dev) {
@@ -2513,13 +2519,13 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 		}
 	}
 
-	bdev = blkdev_get_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_NDELAY, NULL,
-				 NULL);
-	if (IS_ERR(bdev))
-		return PTR_ERR(bdev);
-	sdev = scsi_device_from_queue(bdev->bd_disk->queue);
+	bdev_handle = bdev_open_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_NDELAY,
+				       NULL, NULL);
+	if (IS_ERR(bdev_handle))
+		return PTR_ERR(bdev_handle);
+	sdev = scsi_device_from_queue(bdev_handle->bdev->bd_disk->queue);
 	if (!sdev) {
-		blkdev_put(bdev, NULL);
+		bdev_release(bdev_handle);
 		return -EINVAL;
 	}
 	put_device(&sdev->sdev_gendev);
@@ -2527,8 +2533,8 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 	/* This is safe, since we have a reference from open(). */
 	__module_get(THIS_MODULE);
 
-	pd->bdev = bdev;
-	set_blocksize(bdev, CD_FRAMESIZE);
+	pd->bdev_handle = bdev_handle;
+	set_blocksize(bdev_handle->bdev, CD_FRAMESIZE);
 
 	pkt_init_queue(pd);
 
@@ -2540,11 +2546,11 @@ static int pkt_new_dev(struct pktcdvd_device *pd, dev_t dev)
 	}
 
 	proc_create_single_data(pd->disk->disk_name, 0, pkt_proc, pkt_seq_show, pd);
-	dev_notice(ddev, "writer mapped to %pg\n", bdev);
+	dev_notice(ddev, "writer mapped to %pg\n", bdev_handle->bdev);
 	return 0;
 
 out_mem:
-	blkdev_put(bdev, NULL);
+	bdev_release(bdev_handle);
 	/* This is safe: open() is still holding a reference. */
 	module_put(THIS_MODULE);
 	return -ENOMEM;
@@ -2599,9 +2605,9 @@ static unsigned int pkt_check_events(struct gendisk *disk,
 
 	if (!pd)
 		return 0;
-	if (!pd->bdev)
+	if (!pd->bdev_handle)
 		return 0;
-	attached_disk = pd->bdev->bd_disk;
+	attached_disk = pd->bdev_handle->bdev->bd_disk;
 	if (!attached_disk || !attached_disk->fops->check_events)
 		return 0;
 	return attached_disk->fops->check_events(attached_disk, clearing);
@@ -2686,7 +2692,7 @@ static int pkt_setup_dev(dev_t dev, dev_t* pkt_dev)
 		goto out_mem2;
 
 	/* inherit events of the host device */
-	disk->events = pd->bdev->bd_disk->events;
+	disk->events = pd->bdev_handle->bdev->bd_disk->events;
 
 	ret = add_disk(disk);
 	if (ret)
@@ -2751,7 +2757,7 @@ static int pkt_remove_dev(dev_t pkt_dev)
 	pkt_debugfs_dev_remove(pd);
 	pkt_sysfs_dev_remove(pd);
 
-	blkdev_put(pd->bdev, NULL);
+	bdev_release(pd->bdev_handle);
 
 	remove_proc_entry(pd->disk->disk_name, pkt_proc);
 	dev_notice(ddev, "writer unmapped\n");
@@ -2778,7 +2784,7 @@ static void pkt_get_status(struct pkt_ctrl_command *ctrl_cmd)
 
 	pd = pkt_find_dev_from_minor(ctrl_cmd->dev_index);
 	if (pd) {
-		ctrl_cmd->dev = new_encode_dev(pd->bdev->bd_dev);
+		ctrl_cmd->dev = new_encode_dev(pd->bdev_handle->bdev->bd_dev);
 		ctrl_cmd->pkt_dev = new_encode_dev(pd->pkt_dev);
 	} else {
 		ctrl_cmd->dev = 0;
include/linux/pktcdvd.h +3 -1
index 80cb00db42a4..79594aeb160d 100644
@@ -154,7 +154,9 @@ struct packet_stacked_data
 
 struct pktcdvd_device
 {
-	struct block_device	*bdev;		/* dev attached */
+	struct bdev_handle	*bdev_handle;	/* dev attached */
+	/* handle acquired for bdev during pkt_open_dev() */
+	struct bdev_handle	*open_bdev_handle;
 	dev_t			pkt_dev;	/* our dev */
 	struct packet_settings	settings;
 	struct packet_stats	stats;
Signature · 1 line
2.35.3
author reviewer
#20 lore raw

PATCH v3 6/29 rnbd-srv: Convert to use bdev_open_by_path()

In reply to Jan Kara

Convert rnbd-srv to use bdev_open_by_path() and pass the handle around.

---
 drivers/block/rnbd/rnbd-srv.c | 27 ++++++++++++++-------------
 drivers/block/rnbd/rnbd-srv.h |  2 +-
 2 files changed, 15 insertions(+), 14 deletions(-)
drivers/block/rnbd/rnbd-srv.c +14 -13
index c186df0ec641..65de51f3dfd9 100644
@@ -145,7 +145,7 @@ static int process_rdma(struct rnbd_srv_session *srv_sess,
 	priv->sess_dev = sess_dev;
 	priv->id = id;
 
-	bio = bio_alloc(sess_dev->bdev, 1,
+	bio = bio_alloc(sess_dev->bdev_handle->bdev, 1,
 			rnbd_to_bio_flags(le32_to_cpu(msg->rw)), GFP_KERNEL);
 	if (bio_add_page(bio, virt_to_page(data), datalen,
 			offset_in_page(data)) != datalen) {
@@ -219,7 +219,7 @@ void rnbd_destroy_sess_dev(struct rnbd_srv_sess_dev *sess_dev, bool keep_id)
 	rnbd_put_sess_dev(sess_dev);
 	wait_for_completion(&dc); /* wait for inflights to drop to zero */
 
-	blkdev_put(sess_dev->bdev, NULL);
+	bdev_release(sess_dev->bdev_handle);
 	mutex_lock(&sess_dev->dev->lock);
 	list_del(&sess_dev->dev_list);
 	if (!sess_dev->readonly)
@@ -534,7 +534,7 @@ rnbd_srv_get_or_create_srv_dev(struct block_device *bdev,
 static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 					struct rnbd_srv_sess_dev *sess_dev)
 {
-	struct block_device *bdev = sess_dev->bdev;
+	struct block_device *bdev = sess_dev->bdev_handle->bdev;
 
 	rsp->hdr.type = cpu_to_le16(RNBD_MSG_OPEN_RSP);
 	rsp->device_id = cpu_to_le32(sess_dev->device_id);
@@ -559,7 +559,7 @@ static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 static struct rnbd_srv_sess_dev *
 rnbd_srv_create_set_sess_dev(struct rnbd_srv_session *srv_sess,
 			      const struct rnbd_msg_open *open_msg,
-			      struct block_device *bdev, bool readonly,
+			      struct bdev_handle *handle, bool readonly,
 			      struct rnbd_srv_dev *srv_dev)
 {
 	struct rnbd_srv_sess_dev *sdev = rnbd_sess_dev_alloc(srv_sess);
@@ -571,7 +571,7 @@ rnbd_srv_create_set_sess_dev(struct rnbd_srv_session *srv_sess,
 
 	strscpy(sdev->pathname, open_msg->dev_name, sizeof(sdev->pathname));
 
-	sdev->bdev		= bdev;
+	sdev->bdev_handle	= handle;
 	sdev->sess		= srv_sess;
 	sdev->dev		= srv_dev;
 	sdev->readonly		= readonly;
@@ -676,7 +676,7 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 	struct rnbd_srv_dev *srv_dev;
 	struct rnbd_srv_sess_dev *srv_sess_dev;
 	const struct rnbd_msg_open *open_msg = msg;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	blk_mode_t open_flags = BLK_OPEN_READ;
 	char *full_path;
 	struct rnbd_msg_open_rsp *rsp = data;
@@ -714,15 +714,15 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 		goto reject;
 	}
 
-	bdev = blkdev_get_by_path(full_path, open_flags, NULL, NULL);
-	if (IS_ERR(bdev)) {
-		ret = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_path(full_path, open_flags, NULL, NULL);
+	if (IS_ERR(bdev_handle)) {
+		ret = PTR_ERR(bdev_handle);
 		pr_err("Opening device '%s' on session %s failed, failed to open the block device, err: %d\n",
 		       full_path, srv_sess->sessname, ret);
 		goto free_path;
 	}
 
-	srv_dev = rnbd_srv_get_or_create_srv_dev(bdev, srv_sess,
+	srv_dev = rnbd_srv_get_or_create_srv_dev(bdev_handle->bdev, srv_sess,
 						  open_msg->access_mode);
 	if (IS_ERR(srv_dev)) {
 		pr_err("Opening device '%s' on session %s failed, creating srv_dev failed, err: %ld\n",
@@ -731,7 +731,8 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 		goto blkdev_put;
 	}
 
-	srv_sess_dev = rnbd_srv_create_set_sess_dev(srv_sess, open_msg, bdev,
+	srv_sess_dev = rnbd_srv_create_set_sess_dev(srv_sess, open_msg,
+				bdev_handle,
 				open_msg->access_mode == RNBD_ACCESS_RO,
 				srv_dev);
 	if (IS_ERR(srv_sess_dev)) {
@@ -747,7 +748,7 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 	 */
 	mutex_lock(&srv_dev->lock);
 	if (!srv_dev->dev_kobj.state_in_sysfs) {
-		ret = rnbd_srv_create_dev_sysfs(srv_dev, bdev);
+		ret = rnbd_srv_create_dev_sysfs(srv_dev, bdev_handle->bdev);
 		if (ret) {
 			mutex_unlock(&srv_dev->lock);
 			rnbd_srv_err(srv_sess_dev,
@@ -790,7 +791,7 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 	}
 	rnbd_put_srv_dev(srv_dev);
 blkdev_put:
-	blkdev_put(bdev, NULL);
+	bdev_release(bdev_handle);
 free_path:
 	kfree(full_path);
 reject:
drivers/block/rnbd/rnbd-srv.h +1 -1
index 1027656dedb0..343cc682b617 100644
@@ -46,7 +46,7 @@ struct rnbd_srv_dev {
 struct rnbd_srv_sess_dev {
 	/* Entry inside rnbd_srv_dev struct */
 	struct list_head		dev_list;
-	struct block_device		*bdev;
+	struct bdev_handle		*bdev_handle;
 	struct rnbd_srv_session		*sess;
 	struct rnbd_srv_dev		*dev;
 	struct kobject                  kobj;
Signature · 1 line
2.35.3
maintainer Acked-by on v3 6/29
#21 lore raw

Re: [PATCH 06/29] rnbd-srv: Convert to use bdev_open_by_path()

Quoted 9 lines · Jan Kara

Convert rnbd-srv to use bdev_open_by_path() and pass the handle around.

---

Looks good to me,

author reviewer
#22 lore raw

PATCH v3 6/29 rnbd-srv: Convert to use bdev_open_by_path()

In reply to Jan Kara

Convert rnbd-srv to use bdev_open_by_path() and pass the handle around.

---
 drivers/block/rnbd/rnbd-srv.c | 27 ++++++++++++++-------------
 drivers/block/rnbd/rnbd-srv.h |  2 +-
 2 files changed, 15 insertions(+), 14 deletions(-)
drivers/block/rnbd/rnbd-srv.c +14 -13
index c186df0ec641..65de51f3dfd9 100644
@@ -145,7 +145,7 @@ static int process_rdma(struct rnbd_srv_session *srv_sess,
 	priv->sess_dev = sess_dev;
 	priv->id = id;
 
-	bio = bio_alloc(sess_dev->bdev, 1,
+	bio = bio_alloc(sess_dev->bdev_handle->bdev, 1,
 			rnbd_to_bio_flags(le32_to_cpu(msg->rw)), GFP_KERNEL);
 	if (bio_add_page(bio, virt_to_page(data), datalen,
 			offset_in_page(data)) != datalen) {
@@ -219,7 +219,7 @@ void rnbd_destroy_sess_dev(struct rnbd_srv_sess_dev *sess_dev, bool keep_id)
 	rnbd_put_sess_dev(sess_dev);
 	wait_for_completion(&dc); /* wait for inflights to drop to zero */
 
-	blkdev_put(sess_dev->bdev, NULL);
+	bdev_release(sess_dev->bdev_handle);
 	mutex_lock(&sess_dev->dev->lock);
 	list_del(&sess_dev->dev_list);
 	if (!sess_dev->readonly)
@@ -534,7 +534,7 @@ rnbd_srv_get_or_create_srv_dev(struct block_device *bdev,
 static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 					struct rnbd_srv_sess_dev *sess_dev)
 {
-	struct block_device *bdev = sess_dev->bdev;
+	struct block_device *bdev = sess_dev->bdev_handle->bdev;
 
 	rsp->hdr.type = cpu_to_le16(RNBD_MSG_OPEN_RSP);
 	rsp->device_id = cpu_to_le32(sess_dev->device_id);
@@ -559,7 +559,7 @@ static void rnbd_srv_fill_msg_open_rsp(struct rnbd_msg_open_rsp *rsp,
 static struct rnbd_srv_sess_dev *
 rnbd_srv_create_set_sess_dev(struct rnbd_srv_session *srv_sess,
 			      const struct rnbd_msg_open *open_msg,
-			      struct block_device *bdev, bool readonly,
+			      struct bdev_handle *handle, bool readonly,
 			      struct rnbd_srv_dev *srv_dev)
 {
 	struct rnbd_srv_sess_dev *sdev = rnbd_sess_dev_alloc(srv_sess);
@@ -571,7 +571,7 @@ rnbd_srv_create_set_sess_dev(struct rnbd_srv_session *srv_sess,
 
 	strscpy(sdev->pathname, open_msg->dev_name, sizeof(sdev->pathname));
 
-	sdev->bdev		= bdev;
+	sdev->bdev_handle	= handle;
 	sdev->sess		= srv_sess;
 	sdev->dev		= srv_dev;
 	sdev->readonly		= readonly;
@@ -676,7 +676,7 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 	struct rnbd_srv_dev *srv_dev;
 	struct rnbd_srv_sess_dev *srv_sess_dev;
 	const struct rnbd_msg_open *open_msg = msg;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	blk_mode_t open_flags = BLK_OPEN_READ;
 	char *full_path;
 	struct rnbd_msg_open_rsp *rsp = data;
@@ -714,15 +714,15 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 		goto reject;
 	}
 
-	bdev = blkdev_get_by_path(full_path, open_flags, NULL, NULL);
-	if (IS_ERR(bdev)) {
-		ret = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_path(full_path, open_flags, NULL, NULL);
+	if (IS_ERR(bdev_handle)) {
+		ret = PTR_ERR(bdev_handle);
 		pr_err("Opening device '%s' on session %s failed, failed to open the block device, err: %d\n",
 		       full_path, srv_sess->sessname, ret);
 		goto free_path;
 	}
 
-	srv_dev = rnbd_srv_get_or_create_srv_dev(bdev, srv_sess,
+	srv_dev = rnbd_srv_get_or_create_srv_dev(bdev_handle->bdev, srv_sess,
 						  open_msg->access_mode);
 	if (IS_ERR(srv_dev)) {
 		pr_err("Opening device '%s' on session %s failed, creating srv_dev failed, err: %ld\n",
@@ -731,7 +731,8 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 		goto blkdev_put;
 	}
 
-	srv_sess_dev = rnbd_srv_create_set_sess_dev(srv_sess, open_msg, bdev,
+	srv_sess_dev = rnbd_srv_create_set_sess_dev(srv_sess, open_msg,
+				bdev_handle,
 				open_msg->access_mode == RNBD_ACCESS_RO,
 				srv_dev);
 	if (IS_ERR(srv_sess_dev)) {
@@ -747,7 +748,7 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 	 */
 	mutex_lock(&srv_dev->lock);
 	if (!srv_dev->dev_kobj.state_in_sysfs) {
-		ret = rnbd_srv_create_dev_sysfs(srv_dev, bdev);
+		ret = rnbd_srv_create_dev_sysfs(srv_dev, bdev_handle->bdev);
 		if (ret) {
 			mutex_unlock(&srv_dev->lock);
 			rnbd_srv_err(srv_sess_dev,
@@ -790,7 +791,7 @@ static int process_msg_open(struct rnbd_srv_session *srv_sess,
 	}
 	rnbd_put_srv_dev(srv_dev);
 blkdev_put:
-	blkdev_put(bdev, NULL);
+	bdev_release(bdev_handle);
 free_path:
 	kfree(full_path);
 reject:
drivers/block/rnbd/rnbd-srv.h +1 -1
index 1027656dedb0..343cc682b617 100644
@@ -46,7 +46,7 @@ struct rnbd_srv_dev {
 struct rnbd_srv_sess_dev {
 	/* Entry inside rnbd_srv_dev struct */
 	struct list_head		dev_list;
-	struct block_device		*bdev;
+	struct bdev_handle		*bdev_handle;
 	struct rnbd_srv_session		*sess;
 	struct rnbd_srv_dev		*dev;
 	struct kobject                  kobj;
Signature · 1 line
2.35.3
author reviewer
#23 lore raw

PATCH v3 7/29 xen/blkback: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert xen/blkback to use bdev_open_by_dev() and pass the handle around.

---
 drivers/block/xen-blkback/blkback.c |  4 +--
 drivers/block/xen-blkback/common.h  |  4 +--
 drivers/block/xen-blkback/xenbus.c  | 40 +++++++++++++++--------------
 3 files changed, 25 insertions(+), 23 deletions(-)
drivers/block/xen-blkback/blkback.c +2 -2
index c362f4ad80ab..4defd7f387c7 100644
@@ -465,7 +465,7 @@ static int xen_vbd_translate(struct phys_req *req, struct xen_blkif *blkif,
 	}
 
 	req->dev  = vbd->pdevice;
-	req->bdev = vbd->bdev;
+	req->bdev = vbd->bdev_handle->bdev;
 	rc = 0;
 
  out:
@@ -969,7 +969,7 @@ static int dispatch_discard_io(struct xen_blkif_ring *ring,
 	int err = 0;
 	int status = BLKIF_RSP_OKAY;
 	struct xen_blkif *blkif = ring->blkif;
-	struct block_device *bdev = blkif->vbd.bdev;
+	struct block_device *bdev = blkif->vbd.bdev_handle->bdev;
 	struct phys_req preq;
 
 	xen_blkif_get(blkif);
drivers/block/xen-blkback/common.h +2 -2
index 40f67bfc052d..5ff50e76cee5 100644
@@ -221,7 +221,7 @@ struct xen_vbd {
 	unsigned char		type;
 	/* phys device that this vbd maps to. */
 	u32			pdevice;
-	struct block_device	*bdev;
+	struct bdev_handle	*bdev_handle;
 	/* Cached size parameter. */
 	sector_t		size;
 	unsigned int		flush_support:1;
@@ -360,7 +360,7 @@ struct pending_req {
 };
 
 
-#define vbd_sz(_v)	bdev_nr_sectors((_v)->bdev)
+#define vbd_sz(_v)	bdev_nr_sectors((_v)->bdev_handle->bdev)
 
 #define xen_blkif_get(_b) (atomic_inc(&(_b)->refcnt))
 #define xen_blkif_put(_b)				\
drivers/block/xen-blkback/xenbus.c +21 -19
index bb66178c432b..e34219ea2b05 100644
@@ -81,7 +81,7 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 	int i;
 
 	/* Not ready to connect? */
-	if (!blkif->rings || !blkif->rings[0].irq || !blkif->vbd.bdev)
+	if (!blkif->rings || !blkif->rings[0].irq || !blkif->vbd.bdev_handle)
 		return;
 
 	/* Already connected? */
@@ -99,12 +99,13 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 		return;
 	}
 
-	err = sync_blockdev(blkif->vbd.bdev);
+	err = sync_blockdev(blkif->vbd.bdev_handle->bdev);
 	if (err) {
 		xenbus_dev_error(blkif->be->dev, err, "block flush");
 		return;
 	}
-	invalidate_inode_pages2(blkif->vbd.bdev->bd_inode->i_mapping);
+	invalidate_inode_pages2(
+			blkif->vbd.bdev_handle->bdev->bd_inode->i_mapping);
 
 	for (i = 0; i < blkif->nr_rings; i++) {
 		ring = &blkif->rings[i];
@@ -472,9 +473,9 @@ static void xenvbd_sysfs_delif(struct xenbus_device *dev)
 
 static void xen_vbd_free(struct xen_vbd *vbd)
 {
-	if (vbd->bdev)
-		blkdev_put(vbd->bdev, NULL);
-	vbd->bdev = NULL;
+	if (vbd->bdev_handle)
+		bdev_release(vbd->bdev_handle);
+	vbd->bdev_handle = NULL;
 }
 
 static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
@@ -482,7 +483,7 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 			  int cdrom)
 {
 	struct xen_vbd *vbd;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 
 	vbd = &blkif->vbd;
 	vbd->handle   = handle;
@@ -491,17 +492,17 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 
 	vbd->pdevice  = MKDEV(major, minor);
 
-	bdev = blkdev_get_by_dev(vbd->pdevice, vbd->readonly ?
+	bdev_handle = bdev_open_by_dev(vbd->pdevice, vbd->readonly ?
 				 BLK_OPEN_READ : BLK_OPEN_WRITE, NULL, NULL);
 
-	if (IS_ERR(bdev)) {
+	if (IS_ERR(bdev_handle)) {
 		pr_warn("xen_vbd_create: device %08x could not be opened\n",
 			vbd->pdevice);
 		return -ENOENT;
 	}
 
-	vbd->bdev = bdev;
-	if (vbd->bdev->bd_disk == NULL) {
+	vbd->bdev_handle = bdev_handle;
+	if (vbd->bdev_handle->bdev->bd_disk == NULL) {
 		pr_warn("xen_vbd_create: device %08x doesn't exist\n",
 			vbd->pdevice);
 		xen_vbd_free(vbd);
@@ -509,14 +510,14 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 	}
 	vbd->size = vbd_sz(vbd);
 
-	if (cdrom || disk_to_cdi(vbd->bdev->bd_disk))
+	if (cdrom || disk_to_cdi(vbd->bdev_handle->bdev->bd_disk))
 		vbd->type |= VDISK_CDROM;
-	if (vbd->bdev->bd_disk->flags & GENHD_FL_REMOVABLE)
+	if (vbd->bdev_handle->bdev->bd_disk->flags & GENHD_FL_REMOVABLE)
 		vbd->type |= VDISK_REMOVABLE;
 
-	if (bdev_write_cache(bdev))
+	if (bdev_write_cache(bdev_handle->bdev))
 		vbd->flush_support = true;
-	if (bdev_max_secure_erase_sectors(bdev))
+	if (bdev_max_secure_erase_sectors(bdev_handle->bdev))
 		vbd->discard_secure = true;
 
 	pr_debug("Successful creation of handle=%04x (dom=%u)\n",
@@ -569,7 +570,7 @@ static void xen_blkbk_discard(struct xenbus_transaction xbt, struct backend_info
 	struct xen_blkif *blkif = be->blkif;
 	int err;
 	int state = 0;
-	struct block_device *bdev = be->blkif->vbd.bdev;
+	struct block_device *bdev = be->blkif->vbd.bdev_handle->bdev;
 
 	if (!xenbus_read_unsigned(dev->nodename, "discard-enable", 1))
 		return;
@@ -930,15 +931,16 @@ static void connect(struct backend_info *be)
 		goto abort;
 	}
 	err = xenbus_printf(xbt, dev->nodename, "sector-size", "%lu",
-			    (unsigned long)
-			    bdev_logical_block_size(be->blkif->vbd.bdev));
+			    (unsigned long)bdev_logical_block_size(
+					be->blkif->vbd.bdev_handle->bdev));
 	if (err) {
 		xenbus_dev_fatal(dev, err, "writing %s/sector-size",
 				 dev->nodename);
 		goto abort;
 	}
 	err = xenbus_printf(xbt, dev->nodename, "physical-sector-size", "%u",
-			    bdev_physical_block_size(be->blkif->vbd.bdev));
+			    bdev_physical_block_size(
+					be->blkif->vbd.bdev_handle->bdev));
 	if (err)
 		xenbus_dev_error(dev, err, "writing %s/physical-sector-size",
 				 dev->nodename);
Signature · 1 line
2.35.3
author reviewer
#25 lore raw

PATCH v3 7/29 xen/blkback: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert xen/blkback to use bdev_open_by_dev() and pass the handle around.

---
 drivers/block/xen-blkback/blkback.c |  4 +--
 drivers/block/xen-blkback/common.h  |  4 +--
 drivers/block/xen-blkback/xenbus.c  | 40 +++++++++++++++--------------
 3 files changed, 25 insertions(+), 23 deletions(-)
drivers/block/xen-blkback/blkback.c +2 -2
index c362f4ad80ab..4defd7f387c7 100644
@@ -465,7 +465,7 @@ static int xen_vbd_translate(struct phys_req *req, struct xen_blkif *blkif,
 	}
 
 	req->dev  = vbd->pdevice;
-	req->bdev = vbd->bdev;
+	req->bdev = vbd->bdev_handle->bdev;
 	rc = 0;
 
  out:
@@ -969,7 +969,7 @@ static int dispatch_discard_io(struct xen_blkif_ring *ring,
 	int err = 0;
 	int status = BLKIF_RSP_OKAY;
 	struct xen_blkif *blkif = ring->blkif;
-	struct block_device *bdev = blkif->vbd.bdev;
+	struct block_device *bdev = blkif->vbd.bdev_handle->bdev;
 	struct phys_req preq;
 
 	xen_blkif_get(blkif);
drivers/block/xen-blkback/common.h +2 -2
index 40f67bfc052d..5ff50e76cee5 100644
@@ -221,7 +221,7 @@ struct xen_vbd {
 	unsigned char		type;
 	/* phys device that this vbd maps to. */
 	u32			pdevice;
-	struct block_device	*bdev;
+	struct bdev_handle	*bdev_handle;
 	/* Cached size parameter. */
 	sector_t		size;
 	unsigned int		flush_support:1;
@@ -360,7 +360,7 @@ struct pending_req {
 };
 
 
-#define vbd_sz(_v)	bdev_nr_sectors((_v)->bdev)
+#define vbd_sz(_v)	bdev_nr_sectors((_v)->bdev_handle->bdev)
 
 #define xen_blkif_get(_b) (atomic_inc(&(_b)->refcnt))
 #define xen_blkif_put(_b)				\
drivers/block/xen-blkback/xenbus.c +21 -19
index bb66178c432b..e34219ea2b05 100644
@@ -81,7 +81,7 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 	int i;
 
 	/* Not ready to connect? */
-	if (!blkif->rings || !blkif->rings[0].irq || !blkif->vbd.bdev)
+	if (!blkif->rings || !blkif->rings[0].irq || !blkif->vbd.bdev_handle)
 		return;
 
 	/* Already connected? */
@@ -99,12 +99,13 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 		return;
 	}
 
-	err = sync_blockdev(blkif->vbd.bdev);
+	err = sync_blockdev(blkif->vbd.bdev_handle->bdev);
 	if (err) {
 		xenbus_dev_error(blkif->be->dev, err, "block flush");
 		return;
 	}
-	invalidate_inode_pages2(blkif->vbd.bdev->bd_inode->i_mapping);
+	invalidate_inode_pages2(
+			blkif->vbd.bdev_handle->bdev->bd_inode->i_mapping);
 
 	for (i = 0; i < blkif->nr_rings; i++) {
 		ring = &blkif->rings[i];
@@ -472,9 +473,9 @@ static void xenvbd_sysfs_delif(struct xenbus_device *dev)
 
 static void xen_vbd_free(struct xen_vbd *vbd)
 {
-	if (vbd->bdev)
-		blkdev_put(vbd->bdev, NULL);
-	vbd->bdev = NULL;
+	if (vbd->bdev_handle)
+		bdev_release(vbd->bdev_handle);
+	vbd->bdev_handle = NULL;
 }
 
 static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
@@ -482,7 +483,7 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 			  int cdrom)
 {
 	struct xen_vbd *vbd;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 
 	vbd = &blkif->vbd;
 	vbd->handle   = handle;
@@ -491,17 +492,17 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 
 	vbd->pdevice  = MKDEV(major, minor);
 
-	bdev = blkdev_get_by_dev(vbd->pdevice, vbd->readonly ?
+	bdev_handle = bdev_open_by_dev(vbd->pdevice, vbd->readonly ?
 				 BLK_OPEN_READ : BLK_OPEN_WRITE, NULL, NULL);
 
-	if (IS_ERR(bdev)) {
+	if (IS_ERR(bdev_handle)) {
 		pr_warn("xen_vbd_create: device %08x could not be opened\n",
 			vbd->pdevice);
 		return -ENOENT;
 	}
 
-	vbd->bdev = bdev;
-	if (vbd->bdev->bd_disk == NULL) {
+	vbd->bdev_handle = bdev_handle;
+	if (vbd->bdev_handle->bdev->bd_disk == NULL) {
 		pr_warn("xen_vbd_create: device %08x doesn't exist\n",
 			vbd->pdevice);
 		xen_vbd_free(vbd);
@@ -509,14 +510,14 @@ static int xen_vbd_create(struct xen_blkif *blkif, blkif_vdev_t handle,
 	}
 	vbd->size = vbd_sz(vbd);
 
-	if (cdrom || disk_to_cdi(vbd->bdev->bd_disk))
+	if (cdrom || disk_to_cdi(vbd->bdev_handle->bdev->bd_disk))
 		vbd->type |= VDISK_CDROM;
-	if (vbd->bdev->bd_disk->flags & GENHD_FL_REMOVABLE)
+	if (vbd->bdev_handle->bdev->bd_disk->flags & GENHD_FL_REMOVABLE)
 		vbd->type |= VDISK_REMOVABLE;
 
-	if (bdev_write_cache(bdev))
+	if (bdev_write_cache(bdev_handle->bdev))
 		vbd->flush_support = true;
-	if (bdev_max_secure_erase_sectors(bdev))
+	if (bdev_max_secure_erase_sectors(bdev_handle->bdev))
 		vbd->discard_secure = true;
 
 	pr_debug("Successful creation of handle=%04x (dom=%u)\n",
@@ -569,7 +570,7 @@ static void xen_blkbk_discard(struct xenbus_transaction xbt, struct backend_info
 	struct xen_blkif *blkif = be->blkif;
 	int err;
 	int state = 0;
-	struct block_device *bdev = be->blkif->vbd.bdev;
+	struct block_device *bdev = be->blkif->vbd.bdev_handle->bdev;
 
 	if (!xenbus_read_unsigned(dev->nodename, "discard-enable", 1))
 		return;
@@ -930,15 +931,16 @@ static void connect(struct backend_info *be)
 		goto abort;
 	}
 	err = xenbus_printf(xbt, dev->nodename, "sector-size", "%lu",
-			    (unsigned long)
-			    bdev_logical_block_size(be->blkif->vbd.bdev));
+			    (unsigned long)bdev_logical_block_size(
+					be->blkif->vbd.bdev_handle->bdev));
 	if (err) {
 		xenbus_dev_fatal(dev, err, "writing %s/sector-size",
 				 dev->nodename);
 		goto abort;
 	}
 	err = xenbus_printf(xbt, dev->nodename, "physical-sector-size", "%u",
-			    bdev_physical_block_size(be->blkif->vbd.bdev));
+			    bdev_physical_block_size(
+					be->blkif->vbd.bdev_handle->bdev));
 	if (err)
 		xenbus_dev_error(dev, err, "writing %s/physical-sector-size",
 				 dev->nodename);
Signature · 1 line
2.35.3
author reviewer
#26 lore raw

PATCH v3 8/29 zram: Convert to use bdev_open_by_dev()

In reply to Jan Kara

Convert zram to use bdev_open_by_dev() and pass the handle around.

---
 drivers/block/zram/zram_drv.c | 31 ++++++++++++++-----------------
 drivers/block/zram/zram_drv.h |  2 +-
 2 files changed, 15 insertions(+), 18 deletions(-)
drivers/block/zram/zram_drv.c +14 -17
index 5676e6dd5b16..dfbb102d9c79 100644
@@ -414,17 +414,14 @@ static ssize_t writeback_limit_show(struct device *dev,
 
 static void reset_bdev(struct zram *zram)
 {
-	struct block_device *bdev;
-
 	if (!zram->backing_dev)
 		return;
 
-	bdev = zram->bdev;
-	blkdev_put(bdev, zram);
+	bdev_release(zram->bdev_handle);
 	/* hope filp_close flush all of IO */
 	filp_close(zram->backing_dev, NULL);
 	zram->backing_dev = NULL;
-	zram->bdev = NULL;
+	zram->bdev_handle = NULL;
 	zram->disk->fops = &zram_devops;
 	kvfree(zram->bitmap);
 	zram->bitmap = NULL;
@@ -470,7 +467,7 @@ static ssize_t backing_dev_store(struct device *dev,
 	struct address_space *mapping;
 	unsigned int bitmap_sz;
 	unsigned long nr_pages, *bitmap = NULL;
-	struct block_device *bdev = NULL;
+	struct bdev_handle *bdev_handle = NULL;
 	int err;
 	struct zram *zram = dev_to_zram(dev);
 
@@ -507,11 +504,11 @@ static ssize_t backing_dev_store(struct device *dev,
 		goto out;
 	}
 
-	bdev = blkdev_get_by_dev(inode->i_rdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
-				 zram, NULL);
-	if (IS_ERR(bdev)) {
-		err = PTR_ERR(bdev);
-		bdev = NULL;
+	bdev_handle = bdev_open_by_dev(inode->i_rdev,
+				BLK_OPEN_READ | BLK_OPEN_WRITE, zram, NULL);
+	if (IS_ERR(bdev_handle)) {
+		err = PTR_ERR(bdev_handle);
+		bdev_handle = NULL;
 		goto out;
 	}
 
@@ -525,7 +522,7 @@ static ssize_t backing_dev_store(struct device *dev,
 
 	reset_bdev(zram);
 
-	zram->bdev = bdev;
+	zram->bdev_handle = bdev_handle;
 	zram->backing_dev = backing_dev;
 	zram->bitmap = bitmap;
 	zram->nr_pages = nr_pages;
@@ -538,8 +535,8 @@ static ssize_t backing_dev_store(struct device *dev,
 out:
 	kvfree(bitmap);
 
-	if (bdev)
-		blkdev_put(bdev, zram);
+	if (bdev_handle)
+		bdev_release(bdev_handle);
 
 	if (backing_dev)
 		filp_close(backing_dev, NULL);
@@ -581,7 +578,7 @@ static void read_from_bdev_async(struct zram *zram, struct page *page,
 {
 	struct bio *bio;
 
-	bio = bio_alloc(zram->bdev, 1, parent->bi_opf, GFP_NOIO);
+	bio = bio_alloc(zram->bdev_handle->bdev, 1, parent->bi_opf, GFP_NOIO);
 	bio->bi_iter.bi_sector = entry * (PAGE_SIZE >> 9);
 	__bio_add_page(bio, page, PAGE_SIZE, 0);
 	bio_chain(bio, parent);
@@ -697,7 +694,7 @@ static ssize_t writeback_store(struct device *dev,
 			continue;
 		}
 
-		bio_init(&bio, zram->bdev, &bio_vec, 1,
+		bio_init(&bio, zram->bdev_handle->bdev, &bio_vec, 1,
 			 REQ_OP_WRITE | REQ_SYNC);
 		bio.bi_iter.bi_sector = blk_idx * (PAGE_SIZE >> 9);
 		__bio_add_page(&bio, page, PAGE_SIZE, 0);
@@ -779,7 +776,7 @@ static void zram_sync_read(struct work_struct *work)
 	struct bio_vec bv;
 	struct bio bio;
 
-	bio_init(&bio, zw->zram->bdev, &bv, 1, REQ_OP_READ);
+	bio_init(&bio, zw->zram->bdev_handle->bdev, &bv, 1, REQ_OP_READ);
 	bio.bi_iter.bi_sector = zw->entry * (PAGE_SIZE >> 9);
 	__bio_add_page(&bio, zw->page, PAGE_SIZE, 0);
 	zw->error = submit_bio_wait(&bio);
drivers/block/zram/zram_drv.h +1 -1
index ca7a15bd4845..d090753f97be 100644
@@ -132,7 +132,7 @@ struct zram {
 	spinlock_t wb_limit_lock;
 	bool wb_limit_enable;
 	u64 bd_wb_limit;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	unsigned long *bitmap;
 	unsigned long nr_pages;
 #endif
Signature · 1 line
2.35.3
maintainer Acked-by on v3 8/29
#27 lore raw

Re: [PATCH 08/29] zram: Convert to use bdev_open_by_dev()

Quoted 8 lines · Jan Kara

Convert zram to use bdev_open_by_dev() and pass the handle around.

---

Looks good to me,

author reviewer
#28 lore raw

PATCH v3 8/29 zram: Convert to use bdev_open_by_dev()

In reply to Jan Kara

Convert zram to use bdev_open_by_dev() and pass the handle around.

---
 drivers/block/zram/zram_drv.c | 31 ++++++++++++++-----------------
 drivers/block/zram/zram_drv.h |  2 +-
 2 files changed, 15 insertions(+), 18 deletions(-)
drivers/block/zram/zram_drv.c +14 -17
index 06673c6ca255..d77d3664ca08 100644
@@ -414,17 +414,14 @@ static ssize_t writeback_limit_show(struct device *dev,
 
 static void reset_bdev(struct zram *zram)
 {
-	struct block_device *bdev;
-
 	if (!zram->backing_dev)
 		return;
 
-	bdev = zram->bdev;
-	blkdev_put(bdev, zram);
+	bdev_release(zram->bdev_handle);
 	/* hope filp_close flush all of IO */
 	filp_close(zram->backing_dev, NULL);
 	zram->backing_dev = NULL;
-	zram->bdev = NULL;
+	zram->bdev_handle = NULL;
 	zram->disk->fops = &zram_devops;
 	kvfree(zram->bitmap);
 	zram->bitmap = NULL;
@@ -470,7 +467,7 @@ static ssize_t backing_dev_store(struct device *dev,
 	struct address_space *mapping;
 	unsigned int bitmap_sz;
 	unsigned long nr_pages, *bitmap = NULL;
-	struct block_device *bdev = NULL;
+	struct bdev_handle *bdev_handle = NULL;
 	int err;
 	struct zram *zram = dev_to_zram(dev);
 
@@ -507,11 +504,11 @@ static ssize_t backing_dev_store(struct device *dev,
 		goto out;
 	}
 
-	bdev = blkdev_get_by_dev(inode->i_rdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
-				 zram, NULL);
-	if (IS_ERR(bdev)) {
-		err = PTR_ERR(bdev);
-		bdev = NULL;
+	bdev_handle = bdev_open_by_dev(inode->i_rdev,
+				BLK_OPEN_READ | BLK_OPEN_WRITE, zram, NULL);
+	if (IS_ERR(bdev_handle)) {
+		err = PTR_ERR(bdev_handle);
+		bdev_handle = NULL;
 		goto out;
 	}
 
@@ -525,7 +522,7 @@ static ssize_t backing_dev_store(struct device *dev,
 
 	reset_bdev(zram);
 
-	zram->bdev = bdev;
+	zram->bdev_handle = bdev_handle;
 	zram->backing_dev = backing_dev;
 	zram->bitmap = bitmap;
 	zram->nr_pages = nr_pages;
@@ -538,8 +535,8 @@ static ssize_t backing_dev_store(struct device *dev,
 out:
 	kvfree(bitmap);
 
-	if (bdev)
-		blkdev_put(bdev, zram);
+	if (bdev_handle)
+		bdev_release(bdev_handle);
 
 	if (backing_dev)
 		filp_close(backing_dev, NULL);
@@ -581,7 +578,7 @@ static void read_from_bdev_async(struct zram *zram, struct page *page,
 {
 	struct bio *bio;
 
-	bio = bio_alloc(zram->bdev, 1, parent->bi_opf, GFP_NOIO);
+	bio = bio_alloc(zram->bdev_handle->bdev, 1, parent->bi_opf, GFP_NOIO);
 	bio->bi_iter.bi_sector = entry * (PAGE_SIZE >> 9);
 	__bio_add_page(bio, page, PAGE_SIZE, 0);
 	bio_chain(bio, parent);
@@ -697,7 +694,7 @@ static ssize_t writeback_store(struct device *dev,
 			continue;
 		}
 
-		bio_init(&bio, zram->bdev, &bio_vec, 1,
+		bio_init(&bio, zram->bdev_handle->bdev, &bio_vec, 1,
 			 REQ_OP_WRITE | REQ_SYNC);
 		bio.bi_iter.bi_sector = blk_idx * (PAGE_SIZE >> 9);
 		__bio_add_page(&bio, page, PAGE_SIZE, 0);
@@ -779,7 +776,7 @@ static void zram_sync_read(struct work_struct *work)
 	struct bio_vec bv;
 	struct bio bio;
 
-	bio_init(&bio, zw->zram->bdev, &bv, 1, REQ_OP_READ);
+	bio_init(&bio, zw->zram->bdev_handle->bdev, &bv, 1, REQ_OP_READ);
 	bio.bi_iter.bi_sector = zw->entry * (PAGE_SIZE >> 9);
 	__bio_add_page(&bio, zw->page, PAGE_SIZE, 0);
 	zw->error = submit_bio_wait(&bio);
drivers/block/zram/zram_drv.h +1 -1
index ca7a15bd4845..d090753f97be 100644
@@ -132,7 +132,7 @@ struct zram {
 	spinlock_t wb_limit_lock;
 	bool wb_limit_enable;
 	u64 bd_wb_limit;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	unsigned long *bitmap;
 	unsigned long nr_pages;
 #endif
Signature · 1 line
2.35.3
author reviewer
#29 lore raw

PATCH v3 9/29 bcache: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert bcache to use bdev_open_by_path() and pass the handle around.

---
 drivers/md/bcache/bcache.h |  2 +
 drivers/md/bcache/super.c  | 78 ++++++++++++++++++++------------------
 2 files changed, 43 insertions(+), 37 deletions(-)
drivers/md/bcache/bcache.h +2 -0
index 5a79bb3c272f..2aa3f2c1f719 100644
@@ -299,6 +299,7 @@ struct cached_dev {
 	struct list_head	list;
 	struct bcache_device	disk;
 	struct block_device	*bdev;
+	struct bdev_handle	*bdev_handle;
 
 	struct cache_sb		sb;
 	struct cache_sb_disk	*sb_disk;
@@ -421,6 +422,7 @@ struct cache {
 
 	struct kobject		kobj;
 	struct block_device	*bdev;
+	struct bdev_handle	*bdev_handle;
 
 	struct task_struct	*alloc_thread;
 
drivers/md/bcache/super.c +41 -37

211 lines not on this page · raw message

Signature · 1 line
2.35.3
maintainer Acked-by on v3 9/29
#30 lore raw

Re: [PATCH 09/29] bcache: Convert to bdev_open_by_path()

Quoted 9 lines · Jan Kara

Convert bcache to use bdev_open_by_path() and pass the handle around.

---

Looks good to me,

drivers/md/bcache/super.c static void cached_dev_free(struct closure *cl) · line 1372

Quoted 4 lines
drivers/md/bcache/super.c +2 -2
-	if (!IS_ERR_OR_NULL(dc->bdev))
-		blkdev_put(dc->bdev, dc);
+	if (dc->bdev_handle)
+		bdev_release(dc->bdev_handle);

Fwiw, these conversions confused me a little as the old check gave the impression that this could be set to an error pointer somehow.

author reviewer
#31 lore raw

PATCH v3 9/29 bcache: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert bcache to use bdev_open_by_path() and pass the handle around.

---
 drivers/md/bcache/bcache.h |  2 +
 drivers/md/bcache/super.c  | 78 ++++++++++++++++++++------------------
 2 files changed, 43 insertions(+), 37 deletions(-)
drivers/md/bcache/bcache.h +2 -0
index 5a79bb3c272f..2aa3f2c1f719 100644
@@ -299,6 +299,7 @@ struct cached_dev {
 	struct list_head	list;
 	struct bcache_device	disk;
 	struct block_device	*bdev;
+	struct bdev_handle	*bdev_handle;
 
 	struct cache_sb		sb;
 	struct cache_sb_disk	*sb_disk;
@@ -421,6 +422,7 @@ struct cache {
 
 	struct kobject		kobj;
 	struct block_device	*bdev;
+	struct bdev_handle	*bdev_handle;
 
 	struct task_struct	*alloc_thread;
 
drivers/md/bcache/super.c +41 -37

211 lines not on this page · raw message

Signature · 1 line
2.35.3
author reviewer
#32 lore raw

PATCH v3 10/29 dm: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert device mapper to use bdev_open_by_dev() and pass the handle around.

---
 drivers/md/dm.c               | 20 +++++++++++---------
 include/linux/device-mapper.h |  1 +
 2 files changed, 12 insertions(+), 9 deletions(-)
drivers/md/dm.c +11 -9
index f0f118ab20fa..efde77d34f15 100644
@@ -742,7 +742,7 @@ static struct table_device *open_table_device(struct mapped_device *md,
 		dev_t dev, blk_mode_t mode)
 {
 	struct table_device *td;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	u64 part_off;
 	int r;
 
@@ -751,9 +751,9 @@ static struct table_device *open_table_device(struct mapped_device *md,
 		return ERR_PTR(-ENOMEM);
 	refcount_set(&td->count, 1);
 
-	bdev = blkdev_get_by_dev(dev, mode, _dm_claim_ptr, NULL);
-	if (IS_ERR(bdev)) {
-		r = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_dev(dev, mode, _dm_claim_ptr, NULL);
+	if (IS_ERR(bdev_handle)) {
+		r = PTR_ERR(bdev_handle);
 		goto out_free_td;
 	}
 
@@ -763,20 +763,22 @@ static struct table_device *open_table_device(struct mapped_device *md,
 	 * called.
 	 */
 	if (md->disk->slave_dir) {
-		r = bd_link_disk_holder(bdev, md->disk);
+		r = bd_link_disk_holder(bdev_handle->bdev, md->disk);
 		if (r)
 			goto out_blkdev_put;
 	}
 
 	td->dm_dev.mode = mode;
-	td->dm_dev.bdev = bdev;
-	td->dm_dev.dax_dev = fs_dax_get_by_bdev(bdev, &part_off, NULL, NULL);
+	td->dm_dev.bdev = bdev_handle->bdev;
+	td->dm_dev.bdev_handle = bdev_handle;
+	td->dm_dev.dax_dev = fs_dax_get_by_bdev(bdev_handle->bdev, &part_off,
+						NULL, NULL);
 	format_dev_t(td->dm_dev.name, dev);
 	list_add(&td->list, &md->table_devices);
 	return td;
 
 out_blkdev_put:
-	blkdev_put(bdev, _dm_claim_ptr);
+	bdev_release(bdev_handle);
 out_free_td:
 	kfree(td);
 	return ERR_PTR(r);
@@ -789,7 +791,7 @@ static void close_table_device(struct table_device *td, struct mapped_device *md
 {
 	if (md->disk->slave_dir)
 		bd_unlink_disk_holder(td->dm_dev.bdev, md->disk);
-	blkdev_put(td->dm_dev.bdev, _dm_claim_ptr);
+	bdev_release(td->dm_dev.bdev_handle);
 	put_dax(td->dm_dev.dax_dev);
 	list_del(&td->list);
 	kfree(td);
include/linux/device-mapper.h +1 -0
index 69d0435c7ebb..772ab4d74d94 100644
@@ -165,6 +165,7 @@ void dm_error(const char *message);
 
 struct dm_dev {
 	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct dax_device *dax_dev;
 	blk_mode_t mode;
 	char name[16];
Signature · 1 line
2.35.3
author reviewer
#34 lore raw

PATCH v3 10/29 dm: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert device mapper to use bdev_open_by_dev() and pass the handle around.

---
 drivers/md/dm.c               | 20 +++++++++++---------
 include/linux/device-mapper.h |  1 +
 2 files changed, 12 insertions(+), 9 deletions(-)
drivers/md/dm.c +11 -9
index 64a1f306c96c..f7212e8fc27f 100644
@@ -724,7 +724,7 @@ static struct table_device *open_table_device(struct mapped_device *md,
 		dev_t dev, blk_mode_t mode)
 {
 	struct table_device *td;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	u64 part_off;
 	int r;
 
@@ -733,9 +733,9 @@ static struct table_device *open_table_device(struct mapped_device *md,
 		return ERR_PTR(-ENOMEM);
 	refcount_set(&td->count, 1);
 
-	bdev = blkdev_get_by_dev(dev, mode, _dm_claim_ptr, NULL);
-	if (IS_ERR(bdev)) {
-		r = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_dev(dev, mode, _dm_claim_ptr, NULL);
+	if (IS_ERR(bdev_handle)) {
+		r = PTR_ERR(bdev_handle);
 		goto out_free_td;
 	}
 
@@ -745,20 +745,22 @@ static struct table_device *open_table_device(struct mapped_device *md,
 	 * called.
 	 */
 	if (md->disk->slave_dir) {
-		r = bd_link_disk_holder(bdev, md->disk);
+		r = bd_link_disk_holder(bdev_handle->bdev, md->disk);
 		if (r)
 			goto out_blkdev_put;
 	}
 
 	td->dm_dev.mode = mode;
-	td->dm_dev.bdev = bdev;
-	td->dm_dev.dax_dev = fs_dax_get_by_bdev(bdev, &part_off, NULL, NULL);
+	td->dm_dev.bdev = bdev_handle->bdev;
+	td->dm_dev.bdev_handle = bdev_handle;
+	td->dm_dev.dax_dev = fs_dax_get_by_bdev(bdev_handle->bdev, &part_off,
+						NULL, NULL);
 	format_dev_t(td->dm_dev.name, dev);
 	list_add(&td->list, &md->table_devices);
 	return td;
 
 out_blkdev_put:
-	blkdev_put(bdev, _dm_claim_ptr);
+	bdev_release(bdev_handle);
 out_free_td:
 	kfree(td);
 	return ERR_PTR(r);
@@ -771,7 +773,7 @@ static void close_table_device(struct table_device *td, struct mapped_device *md
 {
 	if (md->disk->slave_dir)
 		bd_unlink_disk_holder(td->dm_dev.bdev, md->disk);
-	blkdev_put(td->dm_dev.bdev, _dm_claim_ptr);
+	bdev_release(td->dm_dev.bdev_handle);
 	put_dax(td->dm_dev.dax_dev);
 	list_del(&td->list);
 	kfree(td);
include/linux/device-mapper.h +1 -0
index 69d0435c7ebb..772ab4d74d94 100644
@@ -165,6 +165,7 @@ void dm_error(const char *message);
 
 struct dm_dev {
 	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct dax_device *dax_dev;
 	blk_mode_t mode;
 	char name[16];
Signature · 1 line
2.35.3
author reviewer
#35 lore raw

PATCH v3 11/29 md: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert md to use bdev_open_by_dev() and pass the handle around.

---
 drivers/md/md.c | 12 +++++++-----
 drivers/md/md.h |  1 +
 2 files changed, 8 insertions(+), 5 deletions(-)
drivers/md/md.c +7 -5
index 78be7811a89f..20d6cefda3e8 100644
@@ -2450,7 +2450,7 @@ static void export_rdev(struct md_rdev *rdev, struct mddev *mddev)
 	if (test_bit(AutoDetected, &rdev->flags))
 		md_autodetect_dev(rdev->bdev->bd_dev);
 #endif
-	blkdev_put(rdev->bdev, mddev->external ? &claim_rdev : rdev);
+	bdev_release(rdev->bdev_handle);
 	rdev->bdev = NULL;
 	kobject_put(&rdev->kobj);
 }
@@ -3644,14 +3644,16 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	if (err)
 		goto out_clear_rdev;
 
-	rdev->bdev = blkdev_get_by_dev(newdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
+	rdev->bdev_handle = bdev_open_by_dev(newdev,
+			BLK_OPEN_READ | BLK_OPEN_WRITE,
 			super_format == -2 ? &claim_rdev : rdev, NULL);
-	if (IS_ERR(rdev->bdev)) {
+	if (IS_ERR(rdev->bdev_handle)) {
 		pr_warn("md: could not open device unknown-block(%u,%u).\n",
 			MAJOR(newdev), MINOR(newdev));
-		err = PTR_ERR(rdev->bdev);
+		err = PTR_ERR(rdev->bdev_handle);
 		goto out_clear_rdev;
 	}
+	rdev->bdev = rdev->bdev_handle->bdev;
 
 	kobject_init(&rdev->kobj, &rdev_ktype);
 
@@ -3682,7 +3684,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	return rdev;
 
 out_blkdev_put:
-	blkdev_put(rdev->bdev, super_format == -2 ? &claim_rdev : rdev);
+	bdev_release(rdev->bdev_handle);
 out_clear_rdev:
 	md_rdev_clear(rdev);
 out_free_rdev:
drivers/md/md.h +1 -0
index 1aef86bf3fc3..e8108845157b 100644
@@ -59,6 +59,7 @@ struct md_rdev {
 	 */
 	struct block_device *meta_bdev;
 	struct block_device *bdev;	/* block device handle */
+	struct bdev_handle *bdev_handle;	/* Handle from open for bdev */
 
 	struct page	*sb_page, *bb_page;
 	int		sb_loaded;
Signature · 1 line
2.35.3
author reviewer
#37 lore raw

PATCH v3 11/29 md: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert md to use bdev_open_by_dev() and pass the handle around. We also don't need the 'Holder' flag anymore so remove it.

---
 drivers/md/md.c | 22 ++++++++--------------
 drivers/md/md.h |  4 +---
 2 files changed, 9 insertions(+), 17 deletions(-)
drivers/md/md.c +8 -14
index a104a025084d..5ed8cad3f4fa 100644
@@ -2452,8 +2452,7 @@ static void export_rdev(struct md_rdev *rdev, struct mddev *mddev)
 	if (test_bit(AutoDetected, &rdev->flags))
 		md_autodetect_dev(rdev->bdev->bd_dev);
 #endif
-	blkdev_put(rdev->bdev,
-		   test_bit(Holder, &rdev->flags) ? rdev : &claim_rdev);
+	bdev_release(rdev->bdev_handle);
 	rdev->bdev = NULL;
 	kobject_put(&rdev->kobj);
 }
@@ -3648,21 +3647,16 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	if (err)
 		goto out_clear_rdev;
 
-	if (super_format == -2) {
-		holder = &claim_rdev;
-	} else {
-		holder = rdev;
-		set_bit(Holder, &rdev->flags);
-	}
-
-	rdev->bdev = blkdev_get_by_dev(newdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
-				       holder, NULL);
-	if (IS_ERR(rdev->bdev)) {
+	rdev->bdev_handle = bdev_open_by_dev(newdev,
+			BLK_OPEN_READ | BLK_OPEN_WRITE,
+			super_format == -2 ? &claim_rdev : rdev, NULL);
+	if (IS_ERR(rdev->bdev_handle)) {
 		pr_warn("md: could not open device unknown-block(%u,%u).\n",
 			MAJOR(newdev), MINOR(newdev));
-		err = PTR_ERR(rdev->bdev);
+		err = PTR_ERR(rdev->bdev_handle);
 		goto out_clear_rdev;
 	}
+	rdev->bdev = rdev->bdev_handle->bdev;
 
 	kobject_init(&rdev->kobj, &rdev_ktype);
 
@@ -3693,7 +3687,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe
 	return rdev;
 
 out_blkdev_put:
-	blkdev_put(rdev->bdev, holder);
+	bdev_release(rdev->bdev_handle);
 out_clear_rdev:
 	md_rdev_clear(rdev);
 out_free_rdev:
drivers/md/md.h +1 -3
index 7c9c13abd7ca..274e7d61d19f 100644
@@ -59,6 +59,7 @@ struct md_rdev {
 	 */
 	struct block_device *meta_bdev;
 	struct block_device *bdev;	/* block device handle */
+	struct bdev_handle *bdev_handle;	/* Handle from open for bdev */
 
 	struct page	*sb_page, *bb_page;
 	int		sb_loaded;
@@ -211,9 +212,6 @@ enum flag_bits {
 				 * check if there is collision between raid1
 				 * serial bios.
 				 */
-	Holder,			/* rdev is used as holder while opening
-				 * underlying disk exclusively.
-				 */
 };
 
 static inline int is_badblock(struct md_rdev *rdev, sector_t s, int sectors,
Signature · 1 line
2.35.3
author reviewer
#38 lore raw

PATCH v3 12/29 mtd: block2mtd: Convert to bdev_open_by_dev/path()

In reply to Jan Kara

Convert block2mtd to use bdev_open_by_dev() and bdev_open_by_path() and pass the handle around.

---
 drivers/mtd/devices/block2mtd.c | 51 +++++++++++++++++++--------------
 1 file changed, 29 insertions(+), 22 deletions(-)
drivers/mtd/devices/block2mtd.c +29 -22
index be106dc20ff3..aa44a23ec045 100644
@@ -37,7 +37,7 @@
 /* Info for the block device */
 struct block2mtd_dev {
 	struct list_head list;
-	struct block_device *blkdev;
+	struct bdev_handle *bdev_handle;
 	struct mtd_info mtd;
 	struct mutex write_mutex;
 };
@@ -55,7 +55,8 @@ static struct page *page_read(struct address_space *mapping, pgoff_t index)
 /* erase a specified part of the device */
 static int _block2mtd_erase(struct block2mtd_dev *dev, loff_t to, size_t len)
 {
-	struct address_space *mapping = dev->blkdev->bd_inode->i_mapping;
+	struct address_space *mapping =
+				dev->bdev_handle->bdev->bd_inode->i_mapping;
 	struct page *page;
 	pgoff_t index = to >> PAGE_SHIFT;	// page index
 	int pages = len >> PAGE_SHIFT;
@@ -105,6 +106,8 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 		size_t *retlen, u_char *buf)
 {
 	struct block2mtd_dev *dev = mtd->priv;
+	struct address_space *mapping =
+				dev->bdev_handle->bdev->bd_inode->i_mapping;
 	struct page *page;
 	pgoff_t index = from >> PAGE_SHIFT;
 	int offset = from & (PAGE_SIZE-1);
@@ -117,7 +120,7 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 			cpylen = len;	// this page
 		len = len - cpylen;
 
-		page = page_read(dev->blkdev->bd_inode->i_mapping, index);
+		page = page_read(mapping, index);
 		if (IS_ERR(page))
 			return PTR_ERR(page);
 
@@ -139,7 +142,8 @@ static int _block2mtd_write(struct block2mtd_dev *dev, const u_char *buf,
 		loff_t to, size_t len, size_t *retlen)
 {
 	struct page *page;
-	struct address_space *mapping = dev->blkdev->bd_inode->i_mapping;
+	struct address_space *mapping =
+				dev->bdev_handle->bdev->bd_inode->i_mapping;
 	pgoff_t index = to >> PAGE_SHIFT;	// page index
 	int offset = to & ~PAGE_MASK;	// page offset
 	int cpylen;
@@ -194,7 +198,7 @@ static int block2mtd_write(struct mtd_info *mtd, loff_t to, size_t len,
 static void block2mtd_sync(struct mtd_info *mtd)
 {
 	struct block2mtd_dev *dev = mtd->priv;
-	sync_blockdev(dev->blkdev);
+	sync_blockdev(dev->bdev_handle->bdev);
 	return;
 }
 
@@ -206,10 +210,10 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 
 	kfree(dev->mtd.name);
 
-	if (dev->blkdev) {
-		invalidate_mapping_pages(dev->blkdev->bd_inode->i_mapping,
-					0, -1);
-		blkdev_put(dev->blkdev, NULL);
+	if (dev->bdev_handle) {
+		invalidate_mapping_pages(
+			dev->bdev_handle->bdev->bd_inode->i_mapping, 0, -1);
+		bdev_release(dev->bdev_handle);
 	}
 
 	kfree(dev);
@@ -219,10 +223,10 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
  * This function is marked __ref because it calls the __init marked
  * early_lookup_bdev when called from the early boot code.
  */
-static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
+static struct bdev_handle __ref *mdtblock_early_get_bdev(const char *devname,
 		blk_mode_t mode, int timeout, struct block2mtd_dev *dev)
 {
-	struct block_device *bdev = ERR_PTR(-ENODEV);
+	struct bdev_handle *bdev_handle = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
 
@@ -230,7 +234,7 @@ static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 	 * We can't use early_lookup_bdev from a running system.
 	 */
 	if (system_state >= SYSTEM_RUNNING)
-		return bdev;
+		return bdev_handle;
 
 	/*
 	 * We might not have the root device mounted at this point.
@@ -249,19 +253,20 @@ static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 		wait_for_device_probe();
 
 		if (!early_lookup_bdev(devname, &devt)) {
-			bdev = blkdev_get_by_dev(devt, mode, dev, NULL);
-			if (!IS_ERR(bdev))
+			bdev_handle = bdev_open_by_dev(devt, mode, dev, NULL);
+			if (!IS_ERR(bdev_handle))
 				break;
 		}
 	}
 #endif
-	return bdev;
+	return bdev_handle;
 }
 
 static struct block2mtd_dev *add_device(char *devname, int erase_size,
 		char *label, int timeout)
 {
 	const blk_mode_t mode = BLK_OPEN_READ | BLK_OPEN_WRITE;
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 	struct block2mtd_dev *dev;
 	char *name;
@@ -274,21 +279,23 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 		return NULL;
 
 	/* Get a handle on the device */
-	bdev = blkdev_get_by_path(devname, mode, dev, NULL);
-	if (IS_ERR(bdev))
-		bdev = mdtblock_early_get_bdev(devname, mode, timeout, dev);
-	if (IS_ERR(bdev)) {
+	bdev_handle = bdev_open_by_path(devname, mode, dev, NULL);
+	if (IS_ERR(bdev_handle))
+		bdev_handle = mdtblock_early_get_bdev(devname, mode, timeout,
+						      dev);
+	if (IS_ERR(bdev_handle)) {
 		pr_err("error: cannot open device %s\n", devname);
 		goto err_free_block2mtd;
 	}
-	dev->blkdev = bdev;
+	dev->bdev_handle = bdev_handle;
+	bdev = bdev_handle->bdev;
 
 	if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) {
 		pr_err("attempting to use an MTD device as a block device\n");
 		goto err_free_block2mtd;
 	}
 
-	if ((long)dev->blkdev->bd_inode->i_size % erase_size) {
+	if ((long)bdev->bd_inode->i_size % erase_size) {
 		pr_err("erasesize must be a divisor of device size\n");
 		goto err_free_block2mtd;
 	}
@@ -306,7 +313,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 
 	dev->mtd.name = name;
 
-	dev->mtd.size = dev->blkdev->bd_inode->i_size & PAGE_MASK;
+	dev->mtd.size = bdev->bd_inode->i_size & PAGE_MASK;
 	dev->mtd.erasesize = erase_size;
 	dev->mtd.writesize = 1;
 	dev->mtd.writebufsize = PAGE_SIZE;
Signature · 6 lines
2.35.3


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
maintainer Acked-by on v3 12/29
#39 lore raw

Re: [PATCH 12/29] mtd: block2mtd: Convert to bdev_open_by_dev/path()

Quoted 8 lines · Jan Kara

Convert block2mtd to use bdev_open_by_dev() and bdev_open_by_path() and pass the handle around.

---

Looks good to me,

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

author reviewer
#40 lore raw

PATCH v3 12/29 mtd: block2mtd: Convert to bdev_open_by_dev/path()

In reply to Jan Kara

Convert block2mtd to use bdev_open_by_dev() and bdev_open_by_path() and pass the handle around.

---
 drivers/mtd/devices/block2mtd.c | 51 +++++++++++++++++++--------------
 1 file changed, 29 insertions(+), 22 deletions(-)
drivers/mtd/devices/block2mtd.c +29 -22
index be106dc20ff3..aa44a23ec045 100644
@@ -37,7 +37,7 @@
 /* Info for the block device */
 struct block2mtd_dev {
 	struct list_head list;
-	struct block_device *blkdev;
+	struct bdev_handle *bdev_handle;
 	struct mtd_info mtd;
 	struct mutex write_mutex;
 };
@@ -55,7 +55,8 @@ static struct page *page_read(struct address_space *mapping, pgoff_t index)
 /* erase a specified part of the device */
 static int _block2mtd_erase(struct block2mtd_dev *dev, loff_t to, size_t len)
 {
-	struct address_space *mapping = dev->blkdev->bd_inode->i_mapping;
+	struct address_space *mapping =
+				dev->bdev_handle->bdev->bd_inode->i_mapping;
 	struct page *page;
 	pgoff_t index = to >> PAGE_SHIFT;	// page index
 	int pages = len >> PAGE_SHIFT;
@@ -105,6 +106,8 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 		size_t *retlen, u_char *buf)
 {
 	struct block2mtd_dev *dev = mtd->priv;
+	struct address_space *mapping =
+				dev->bdev_handle->bdev->bd_inode->i_mapping;
 	struct page *page;
 	pgoff_t index = from >> PAGE_SHIFT;
 	int offset = from & (PAGE_SIZE-1);
@@ -117,7 +120,7 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 			cpylen = len;	// this page
 		len = len - cpylen;
 
-		page = page_read(dev->blkdev->bd_inode->i_mapping, index);
+		page = page_read(mapping, index);
 		if (IS_ERR(page))
 			return PTR_ERR(page);
 
@@ -139,7 +142,8 @@ static int _block2mtd_write(struct block2mtd_dev *dev, const u_char *buf,
 		loff_t to, size_t len, size_t *retlen)
 {
 	struct page *page;
-	struct address_space *mapping = dev->blkdev->bd_inode->i_mapping;
+	struct address_space *mapping =
+				dev->bdev_handle->bdev->bd_inode->i_mapping;
 	pgoff_t index = to >> PAGE_SHIFT;	// page index
 	int offset = to & ~PAGE_MASK;	// page offset
 	int cpylen;
@@ -194,7 +198,7 @@ static int block2mtd_write(struct mtd_info *mtd, loff_t to, size_t len,
 static void block2mtd_sync(struct mtd_info *mtd)
 {
 	struct block2mtd_dev *dev = mtd->priv;
-	sync_blockdev(dev->blkdev);
+	sync_blockdev(dev->bdev_handle->bdev);
 	return;
 }
 
@@ -206,10 +210,10 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 
 	kfree(dev->mtd.name);
 
-	if (dev->blkdev) {
-		invalidate_mapping_pages(dev->blkdev->bd_inode->i_mapping,
-					0, -1);
-		blkdev_put(dev->blkdev, NULL);
+	if (dev->bdev_handle) {
+		invalidate_mapping_pages(
+			dev->bdev_handle->bdev->bd_inode->i_mapping, 0, -1);
+		bdev_release(dev->bdev_handle);
 	}
 
 	kfree(dev);
@@ -219,10 +223,10 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
  * This function is marked __ref because it calls the __init marked
  * early_lookup_bdev when called from the early boot code.
  */
-static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
+static struct bdev_handle __ref *mdtblock_early_get_bdev(const char *devname,
 		blk_mode_t mode, int timeout, struct block2mtd_dev *dev)
 {
-	struct block_device *bdev = ERR_PTR(-ENODEV);
+	struct bdev_handle *bdev_handle = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
 
@@ -230,7 +234,7 @@ static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 	 * We can't use early_lookup_bdev from a running system.
 	 */
 	if (system_state >= SYSTEM_RUNNING)
-		return bdev;
+		return bdev_handle;
 
 	/*
 	 * We might not have the root device mounted at this point.
@@ -249,19 +253,20 @@ static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 		wait_for_device_probe();
 
 		if (!early_lookup_bdev(devname, &devt)) {
-			bdev = blkdev_get_by_dev(devt, mode, dev, NULL);
-			if (!IS_ERR(bdev))
+			bdev_handle = bdev_open_by_dev(devt, mode, dev, NULL);
+			if (!IS_ERR(bdev_handle))
 				break;
 		}
 	}
 #endif
-	return bdev;
+	return bdev_handle;
 }
 
 static struct block2mtd_dev *add_device(char *devname, int erase_size,
 		char *label, int timeout)
 {
 	const blk_mode_t mode = BLK_OPEN_READ | BLK_OPEN_WRITE;
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 	struct block2mtd_dev *dev;
 	char *name;
@@ -274,21 +279,23 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 		return NULL;
 
 	/* Get a handle on the device */
-	bdev = blkdev_get_by_path(devname, mode, dev, NULL);
-	if (IS_ERR(bdev))
-		bdev = mdtblock_early_get_bdev(devname, mode, timeout, dev);
-	if (IS_ERR(bdev)) {
+	bdev_handle = bdev_open_by_path(devname, mode, dev, NULL);
+	if (IS_ERR(bdev_handle))
+		bdev_handle = mdtblock_early_get_bdev(devname, mode, timeout,
+						      dev);
+	if (IS_ERR(bdev_handle)) {
 		pr_err("error: cannot open device %s\n", devname);
 		goto err_free_block2mtd;
 	}
-	dev->blkdev = bdev;
+	dev->bdev_handle = bdev_handle;
+	bdev = bdev_handle->bdev;
 
 	if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) {
 		pr_err("attempting to use an MTD device as a block device\n");
 		goto err_free_block2mtd;
 	}
 
-	if ((long)dev->blkdev->bd_inode->i_size % erase_size) {
+	if ((long)bdev->bd_inode->i_size % erase_size) {
 		pr_err("erasesize must be a divisor of device size\n");
 		goto err_free_block2mtd;
 	}
@@ -306,7 +313,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 
 	dev->mtd.name = name;
 
-	dev->mtd.size = dev->blkdev->bd_inode->i_size & PAGE_MASK;
+	dev->mtd.size = bdev->bd_inode->i_size & PAGE_MASK;
 	dev->mtd.erasesize = erase_size;
 	dev->mtd.writesize = 1;
 	dev->mtd.writebufsize = PAGE_SIZE;
Signature · 6 lines
2.35.3


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
author reviewer
#41 lore raw

PATCH v3 13/29 nvmet: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert nvmet to use bdev_open_by_path() and pass the handle around.

---
 drivers/nvme/target/io-cmd-bdev.c | 20 +++++++++++---------
 drivers/nvme/target/nvmet.h       |  1 +
 2 files changed, 12 insertions(+), 9 deletions(-)
drivers/nvme/target/io-cmd-bdev.c +11 -9
index 2733e0158585..0bf05c19d8b5 100644
@@ -50,9 +50,10 @@ void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id)
 
 void nvmet_bdev_ns_disable(struct nvmet_ns *ns)
 {
-	if (ns->bdev) {
-		blkdev_put(ns->bdev, NULL);
+	if (ns->bdev_handle) {
+		bdev_release(ns->bdev_handle);
 		ns->bdev = NULL;
+		ns->bdev_handle = NULL;
 	}
 }
 
@@ -84,17 +85,18 @@ int nvmet_bdev_ns_enable(struct nvmet_ns *ns)
 	if (ns->buffered_io)
 		return -ENOTBLK;
 
-	ns->bdev = blkdev_get_by_path(ns->device_path,
-			BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL);
-	if (IS_ERR(ns->bdev)) {
-		ret = PTR_ERR(ns->bdev);
+	ns->bdev_handle = bdev_open_by_path(ns->device_path,
+				BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL);
+	if (IS_ERR(ns->bdev_handle)) {
+		ret = PTR_ERR(ns->bdev_handle);
 		if (ret != -ENOTBLK) {
-			pr_err("failed to open block device %s: (%ld)\n",
-					ns->device_path, PTR_ERR(ns->bdev));
+			pr_err("failed to open block device %s: (%d)\n",
+					ns->device_path, ret);
 		}
-		ns->bdev = NULL;
+		ns->bdev_handle = NULL;
 		return ret;
 	}
+	ns->bdev = ns->bdev_handle->bdev;
 	ns->size = bdev_nr_bytes(ns->bdev);
 	ns->blksize_shift = blksize_bits(bdev_logical_block_size(ns->bdev));
 
drivers/nvme/target/nvmet.h +1 -0
index 8cfd60f3b564..360e385be33b 100644
@@ -58,6 +58,7 @@
 
 struct nvmet_ns {
 	struct percpu_ref	ref;
+	struct bdev_handle	*bdev_handle;
 	struct block_device	*bdev;
 	struct file		*file;
 	bool			readonly;
Signature · 1 line
2.35.3
author reviewer
#44 lore raw

PATCH v3 13/29 nvmet: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert nvmet to use bdev_open_by_path() and pass the handle around.

---
 drivers/nvme/target/io-cmd-bdev.c | 20 +++++++++++---------
 drivers/nvme/target/nvmet.h       |  1 +
 2 files changed, 12 insertions(+), 9 deletions(-)
drivers/nvme/target/io-cmd-bdev.c +11 -9
index 468833675cc9..f11400a908f2 100644
@@ -50,9 +50,10 @@ void nvmet_bdev_set_limits(struct block_device *bdev, struct nvme_id_ns *id)
 
 void nvmet_bdev_ns_disable(struct nvmet_ns *ns)
 {
-	if (ns->bdev) {
-		blkdev_put(ns->bdev, NULL);
+	if (ns->bdev_handle) {
+		bdev_release(ns->bdev_handle);
 		ns->bdev = NULL;
+		ns->bdev_handle = NULL;
 	}
 }
 
@@ -84,17 +85,18 @@ int nvmet_bdev_ns_enable(struct nvmet_ns *ns)
 	if (ns->buffered_io)
 		return -ENOTBLK;
 
-	ns->bdev = blkdev_get_by_path(ns->device_path,
-			BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL);
-	if (IS_ERR(ns->bdev)) {
-		ret = PTR_ERR(ns->bdev);
+	ns->bdev_handle = bdev_open_by_path(ns->device_path,
+				BLK_OPEN_READ | BLK_OPEN_WRITE, NULL, NULL);
+	if (IS_ERR(ns->bdev_handle)) {
+		ret = PTR_ERR(ns->bdev_handle);
 		if (ret != -ENOTBLK) {
-			pr_err("failed to open block device %s: (%ld)\n",
-					ns->device_path, PTR_ERR(ns->bdev));
+			pr_err("failed to open block device %s: (%d)\n",
+					ns->device_path, ret);
 		}
-		ns->bdev = NULL;
+		ns->bdev_handle = NULL;
 		return ret;
 	}
+	ns->bdev = ns->bdev_handle->bdev;
 	ns->size = bdev_nr_bytes(ns->bdev);
 	ns->blksize_shift = blksize_bits(bdev_logical_block_size(ns->bdev));
 
drivers/nvme/target/nvmet.h +1 -0
index 8cfd60f3b564..360e385be33b 100644
@@ -58,6 +58,7 @@
 
 struct nvmet_ns {
 	struct percpu_ref	ref;
+	struct bdev_handle	*bdev_handle;
 	struct block_device	*bdev;
 	struct file		*file;
 	bool			readonly;
Signature · 1 line
2.35.3
author reviewer
#45 lore raw

PATCH v3 14/29 s390/dasd: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert dasd to use bdev_open_by_path() and pass the handle around.

---
 drivers/s390/block/dasd.c       | 12 +++++----
 drivers/s390/block/dasd_genhd.c | 45 ++++++++++++++++-----------------
 drivers/s390/block/dasd_int.h   |  2 +-
 drivers/s390/block/dasd_ioctl.c |  2 +-
 4 files changed, 31 insertions(+), 30 deletions(-)
drivers/s390/block/dasd.c +7 -5
index 215597f73be4..16a2d631a169 100644
@@ -412,7 +412,8 @@ dasd_state_ready_to_online(struct dasd_device * device)
 					KOBJ_CHANGE);
 			return 0;
 		}
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
 	}
 	return 0;
 }
@@ -432,7 +433,8 @@ static int dasd_state_online_to_ready(struct dasd_device *device)
 
 	device->state = DASD_STATE_READY;
 	if (device->block && !(device->features & DASD_FEATURE_USERAW))
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
 	return 0;
 }
 
@@ -3590,7 +3592,7 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
 	 * in the other openers.
 	 */
 	if (device->block) {
-		max_count = device->block->bdev ? 0 : -1;
+		max_count = device->block->bdev_handle ? 0 : -1;
 		open_count = atomic_read(&device->block->open_count);
 		if (open_count > max_count) {
 			if (open_count > 0)
@@ -3636,8 +3638,8 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
 		 * so sync bdev first and then wait for our queues to become
 		 * empty
 		 */
-		if (device->block)
-			bdev_mark_dead(device->block->bdev, false);
+		if (device->block && device->block->bdev_handle) {
+			bdev_mark_dead(device->block->bdev_handle->bdev, false);
 		dasd_schedule_device_bh(device);
 		rc = wait_event_interruptible(shutdown_waitq,
 					      _wait_for_empty_queues(device));
drivers/s390/block/dasd_genhd.c +22 -23

74 lines not on this page · raw message

drivers/s390/block/dasd_int.h +1 -1
index 0aa56351da72..73c5eb0ae6ad 100644
@@ -646,7 +646,7 @@ struct dasd_block {
 	struct gendisk *gdp;
 	spinlock_t request_queue_lock;
 	struct blk_mq_tag_set tag_set;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	atomic_t open_count;
 
 	unsigned long blocks;	   /* size of volume in blocks */
drivers/s390/block/dasd_ioctl.c +1 -1
index d55862605b82..61b9675e2a67 100644
@@ -537,7 +537,7 @@ static int __dasd_ioctl_information(struct dasd_block *block,
 	 * This must be hidden from user-space.
 	 */
 	dasd_info->open_count = atomic_read(&block->open_count);
-	if (!block->bdev)
+	if (!block->bdev_handle)
 		dasd_info->open_count++;
 
 	/*
Signature · 1 line
2.35.3
maintainer Acked-by on v3 14/29
#47 lore raw

Re: [PATCH 14/29] s390/dasd: Convert to bdev_open_by_path()

In reply to Jan Kara

Am 23.08.23 um 12:48 schrieb Jan Kara:

Quoted 8 lines

Convert dasd to use bdev_open_by_path() and pass the handle around.

---

The DASD part does not compile. please see below.

Beside of this the patch looks OK to me.

with the error fixed:

Quoted 40 lines
  drivers/s390/block/dasd.c       | 12 +++++----
  drivers/s390/block/dasd_genhd.c | 45 ++++++++++++++++-----------------
  drivers/s390/block/dasd_int.h   |  2 +-
  drivers/s390/block/dasd_ioctl.c |  2 +-
  4 files changed, 31 insertions(+), 30 deletions(-)
drivers/s390/block/dasd.c +5 -3
index 215597f73be4..16a2d631a169 100644
@@ -412,7 +412,8 @@ dasd_state_ready_to_online(struct dasd_device * device)
  					KOBJ_CHANGE);
  			return 0;
  		}
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
  	}
  	return 0;
  }
@@ -432,7 +433,8 @@ static int dasd_state_online_to_ready(struct dasd_device *device)
 
  	device->state = DASD_STATE_READY;
  	if (device->block && !(device->features & DASD_FEATURE_USERAW))
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
  	return 0;
  }
 
@@ -3590,7 +3592,7 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
  	 * in the other openers.
  	 */
  	if (device->block) {
-		max_count = device->block->bdev ? 0 : -1;
+		max_count = device->block->bdev_handle ? 0 : -1;
  		open_count = atomic_read(&device->block->open_count);
  		if (open_count > max_count) {
  			if (open_count > 0)

drivers/s390/block/dasd.c int dasd_generic_set_offline(struct ccw_device *cdev) · line 3641

Quoted 6 lines
drivers/s390/block/dasd.c +1 -2
@@ -3636,8 +3638,8 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
  		 * so sync bdev first and then wait for our queues to become
  		 * empty
  		 */
-		if (device->block)
-			bdev_mark_dead(device->block->bdev, false);
+		if (device->block && device->block->bdev_handle) {

the brace is not needed here and there is no matching right brace.

Quoted 5 lines
drivers/s390/block/dasd.c +1 -0
+			bdev_mark_dead(device->block->bdev_handle->bdev, false);
  		dasd_schedule_device_bh(device);
  		rc = wait_event_interruptible(shutdown_waitq,
  					      _wait_for_empty_queues(device));
author reviewer
#48 lore raw

Re: [PATCH 14/29] s390/dasd: Convert to bdev_open_by_path()

Quoted 10 lines · Stefan Haberland

Am 23.08.23 um 12:48 schrieb Jan Kara:

Quoted 8 lines

Convert dasd to use bdev_open_by_path() and pass the handle around.

---

Quoted 6 lines

The DASD part does not compile. please see below.

Beside of this the patch looks OK to me.

with the error fixed:

Thanks for noticing. I can see Christian has already fixed up the problem in his tree. I guess he'll pick up your ack once he returns from vacation.

Honza

Quoted 54 lines
Quoted 39 lines
  drivers/s390/block/dasd.c       | 12 +++++----
  drivers/s390/block/dasd_genhd.c | 45 ++++++++++++++++-----------------
  drivers/s390/block/dasd_int.h   |  2 +-
  drivers/s390/block/dasd_ioctl.c |  2 +-
  4 files changed, 31 insertions(+), 30 deletions(-)
drivers/s390/block/dasd.c +5 -3
index 215597f73be4..16a2d631a169 100644
@@ -412,7 +412,8 @@ dasd_state_ready_to_online(struct dasd_device * device)
  					KOBJ_CHANGE);
  			return 0;
  		}
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
  	}
  	return 0;
  }
@@ -432,7 +433,8 @@ static int dasd_state_online_to_ready(struct dasd_device *device)
  	device->state = DASD_STATE_READY;
  	if (device->block && !(device->features & DASD_FEATURE_USERAW))
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
  	return 0;
  }
@@ -3590,7 +3592,7 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
  	 * in the other openers.
  	 */
  	if (device->block) {
-		max_count = device->block->bdev ? 0 : -1;
+		max_count = device->block->bdev_handle ? 0 : -1;
  		open_count = atomic_read(&device->block->open_count);
  		if (open_count > max_count) {
  			if (open_count > 0)

drivers/s390/block/dasd.c int dasd_generic_set_offline(struct ccw_device *cdev) · line 3641

Quoted 6 lines
drivers/s390/block/dasd.c +1 -2
@@ -3636,8 +3638,8 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
  		 * so sync bdev first and then wait for our queues to become
  		 * empty
  		 */
-		if (device->block)
-			bdev_mark_dead(device->block->bdev, false);
+		if (device->block && device->block->bdev_handle) {

the brace is not needed here and there is no matching right brace.

Quoted 5 lines
drivers/s390/block/dasd.c +1 -0
+			bdev_mark_dead(device->block->bdev_handle->bdev, false);
  		dasd_schedule_device_bh(device);
  		rc = wait_event_interruptible(shutdown_waitq,
  					      _wait_for_empty_queues(device));
Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
author reviewer
#49 lore raw

PATCH v3 14/29 s390/dasd: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert dasd to use bdev_open_by_path() and pass the handle around.

---
 drivers/s390/block/dasd.c       | 12 +++++----
 drivers/s390/block/dasd_genhd.c | 45 ++++++++++++++++-----------------
 drivers/s390/block/dasd_int.h   |  2 +-
 drivers/s390/block/dasd_ioctl.c |  2 +-
 4 files changed, 31 insertions(+), 30 deletions(-)
drivers/s390/block/dasd.c +7 -5
index 215597f73be4..16a2d631a169 100644
@@ -412,7 +412,8 @@ dasd_state_ready_to_online(struct dasd_device * device)
 					KOBJ_CHANGE);
 			return 0;
 		}
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
 	}
 	return 0;
 }
@@ -432,7 +433,8 @@ static int dasd_state_online_to_ready(struct dasd_device *device)
 
 	device->state = DASD_STATE_READY;
 	if (device->block && !(device->features & DASD_FEATURE_USERAW))
-		disk_uevent(device->block->bdev->bd_disk, KOBJ_CHANGE);
+		disk_uevent(device->block->bdev_handle->bdev->bd_disk,
+			    KOBJ_CHANGE);
 	return 0;
 }
 
@@ -3590,7 +3592,7 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
 	 * in the other openers.
 	 */
 	if (device->block) {
-		max_count = device->block->bdev ? 0 : -1;
+		max_count = device->block->bdev_handle ? 0 : -1;
 		open_count = atomic_read(&device->block->open_count);
 		if (open_count > max_count) {
 			if (open_count > 0)
@@ -3636,8 +3638,8 @@ int dasd_generic_set_offline(struct ccw_device *cdev)
 		 * so sync bdev first and then wait for our queues to become
 		 * empty
 		 */
-		if (device->block)
-			bdev_mark_dead(device->block->bdev, false);
+		if (device->block && device->block->bdev_handle) {
+			bdev_mark_dead(device->block->bdev_handle->bdev, false);
 		dasd_schedule_device_bh(device);
 		rc = wait_event_interruptible(shutdown_waitq,
 					      _wait_for_empty_queues(device));
drivers/s390/block/dasd_genhd.c +22 -23

74 lines not on this page · raw message

drivers/s390/block/dasd_int.h +1 -1
index 8a4dbe9d7741..2e663131adaf 100644
@@ -650,7 +650,7 @@ struct dasd_block {
 	struct gendisk *gdp;
 	spinlock_t request_queue_lock;
 	struct blk_mq_tag_set tag_set;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	atomic_t open_count;
 
 	unsigned long blocks;	   /* size of volume in blocks */
drivers/s390/block/dasd_ioctl.c +1 -1
index d55862605b82..61b9675e2a67 100644
@@ -537,7 +537,7 @@ static int __dasd_ioctl_information(struct dasd_block *block,
 	 * This must be hidden from user-space.
 	 */
 	dasd_info->open_count = atomic_read(&block->open_count);
-	if (!block->bdev)
+	if (!block->bdev_handle)
 		dasd_info->open_count++;
 
 	/*
Signature · 1 line
2.35.3
author reviewer
#50 lore raw

PATCH v3 15/29 scsi: target: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert iblock and pscsi drivers to use bdev_open_by_path() and pass the handle around.

---
 drivers/target/target_core_iblock.c | 19 +++++++++++--------
 drivers/target/target_core_iblock.h |  1 +
 drivers/target/target_core_pscsi.c  | 26 +++++++++++++-------------
 drivers/target/target_core_pscsi.h  |  2 +-
 4 files changed, 26 insertions(+), 22 deletions(-)
drivers/target/target_core_iblock.c +11 -8
index 3d1b511ea284..3695a9f8d8a5 100644
@@ -91,7 +91,8 @@ static int iblock_configure_device(struct se_device *dev)
 {
 	struct iblock_dev *ib_dev = IBLOCK_DEV(dev);
 	struct request_queue *q;
-	struct block_device *bd = NULL;
+	struct bdev_handle *bdev_handle;
+	struct block_device *bd;
 	struct blk_integrity *bi;
 	blk_mode_t mode = BLK_OPEN_READ;
 	unsigned int max_write_zeroes_sectors;
@@ -116,12 +117,14 @@ static int iblock_configure_device(struct se_device *dev)
 	else
 		dev->dev_flags |= DF_READ_ONLY;
 
-	bd = blkdev_get_by_path(ib_dev->ibd_udev_path, mode, ib_dev, NULL);
-	if (IS_ERR(bd)) {
-		ret = PTR_ERR(bd);
+	bdev_handle = bdev_open_by_path(ib_dev->ibd_udev_path, mode, ib_dev,
+					NULL);
+	if (IS_ERR(bdev_handle)) {
+		ret = PTR_ERR(bdev_handle);
 		goto out_free_bioset;
 	}
-	ib_dev->ibd_bd = bd;
+	ib_dev->ibd_bdev_handle = bdev_handle;
+	ib_dev->ibd_bd = bd = bdev_handle->bdev;
 
 	q = bdev_get_queue(bd);
 
@@ -177,7 +180,7 @@ static int iblock_configure_device(struct se_device *dev)
 	return 0;
 
 out_blkdev_put:
-	blkdev_put(ib_dev->ibd_bd, ib_dev);
+	bdev_release(ib_dev->ibd_bdev_handle);
 out_free_bioset:
 	bioset_exit(&ib_dev->ibd_bio_set);
 out:
@@ -202,8 +205,8 @@ static void iblock_destroy_device(struct se_device *dev)
 {
 	struct iblock_dev *ib_dev = IBLOCK_DEV(dev);
 
-	if (ib_dev->ibd_bd != NULL)
-		blkdev_put(ib_dev->ibd_bd, ib_dev);
+	if (ib_dev->ibd_bdev_handle)
+		bdev_release(ib_dev->ibd_bdev_handle);
 	bioset_exit(&ib_dev->ibd_bio_set);
 }
 
drivers/target/target_core_iblock.h +1 -0
index 8c55375d2f75..683f9a55945b 100644
@@ -32,6 +32,7 @@ struct iblock_dev {
 	u32	ibd_flags;
 	struct bio_set	ibd_bio_set;
 	struct block_device *ibd_bd;
+	struct bdev_handle *ibd_bdev_handle;
 	bool ibd_readonly;
 	struct iblock_dev_plug *ibd_plug;
 } ____cacheline_aligned;
drivers/target/target_core_pscsi.c +13 -13
index 0d4f09693ef4..41b7489d37ce 100644
@@ -352,7 +352,7 @@ static int pscsi_create_type_disk(struct se_device *dev, struct scsi_device *sd)
 	struct pscsi_hba_virt *phv = dev->se_hba->hba_ptr;
 	struct pscsi_dev_virt *pdv = PSCSI_DEV(dev);
 	struct Scsi_Host *sh = sd->host;
-	struct block_device *bd;
+	struct bdev_handle *bdev_handle;
 	int ret;
 
 	if (scsi_device_get(sd)) {
@@ -366,18 +366,18 @@ static int pscsi_create_type_disk(struct se_device *dev, struct scsi_device *sd)
 	 * Claim exclusive struct block_device access to struct scsi_device
 	 * for TYPE_DISK and TYPE_ZBC using supplied udev_path
 	 */
-	bd = blkdev_get_by_path(dev->udev_path, BLK_OPEN_WRITE | BLK_OPEN_READ,
-				pdv, NULL);
-	if (IS_ERR(bd)) {
-		pr_err("pSCSI: blkdev_get_by_path() failed\n");
+	bdev_handle = bdev_open_by_path(dev->udev_path,
+				BLK_OPEN_WRITE | BLK_OPEN_READ, pdv, NULL);
+	if (IS_ERR(bdev_handle)) {
+		pr_err("pSCSI: bdev_open_by_path() failed\n");
 		scsi_device_put(sd);
-		return PTR_ERR(bd);
+		return PTR_ERR(bdev_handle);
 	}
-	pdv->pdv_bd = bd;
+	pdv->pdv_bdev_handle = bdev_handle;
 
 	ret = pscsi_add_device_to_list(dev, sd);
 	if (ret) {
-		blkdev_put(pdv->pdv_bd, pdv);
+		bdev_release(bdev_handle);
 		scsi_device_put(sd);
 		return ret;
 	}
@@ -564,9 +564,9 @@ static void pscsi_destroy_device(struct se_device *dev)
 		 * from pscsi_create_type_disk()
 		 */
 		if ((sd->type == TYPE_DISK || sd->type == TYPE_ZBC) &&
-		    pdv->pdv_bd) {
-			blkdev_put(pdv->pdv_bd, pdv);
-			pdv->pdv_bd = NULL;
+		    pdv->pdv_bdev_handle) {
+			bdev_release(pdv->pdv_bdev_handle);
+			pdv->pdv_bdev_handle = NULL;
 		}
 		/*
 		 * For HBA mode PHV_LLD_SCSI_HOST_NO, release the reference
@@ -994,8 +994,8 @@ static sector_t pscsi_get_blocks(struct se_device *dev)
 {
 	struct pscsi_dev_virt *pdv = PSCSI_DEV(dev);
 
-	if (pdv->pdv_bd)
-		return bdev_nr_sectors(pdv->pdv_bd);
+	if (pdv->pdv_bdev_handle)
+		return bdev_nr_sectors(pdv->pdv_bdev_handle->bdev);
 	return 0;
 }
 
drivers/target/target_core_pscsi.h +1 -1
index 23d9a6e340d4..b0a3ef136592 100644
@@ -37,7 +37,7 @@ struct pscsi_dev_virt {
 	int	pdv_channel_id;
 	int	pdv_target_id;
 	int	pdv_lun_id;
-	struct block_device *pdv_bd;
+	struct bdev_handle *pdv_bdev_handle;
 	struct scsi_device *pdv_sd;
 	struct Scsi_Host *pdv_lld_host;
 } ____cacheline_aligned;
Signature · 1 line
2.35.3
#52 lore raw
In reply to Jan Kara
Quoted 15 lines · Jan Kara

Hello,

this is a v3 of the patch series which implements the idea of blkdev_get_by_*() calls returning bdev_handle which is then passed to blkdev_put() [1]. This makes the get and put calls for bdevs more obviously matching and allows us to propagate context from get to put without having to modify all the users (again!). In particular I need to propagate used open flags to blkdev_put() to be able count writeable opens and add support for blocking writes to mounted block devices. I'll send that series separately.

The series is based on Christian's vfs tree as of today as there is quite some overlap. Patches have passed some reasonable testing - I've tested block changes, md, dm, bcache, xfs, btrfs, ext4, swap. More testing or review is always welcome. Thanks! I've pushed out the full branch to:

Quoted 6 lines

git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs.git bdev_handle

to ease review / testing. Since there were not many comments for v2 and Christoph has acked the series I think we should start discussing how to merge the series. Most collisions with this series seem to happen in the filesystems area so VFS tree would seem as the least painful way to merge this. Jens,

I really do like this series especially struct bdev_handle and moving the mode bits in there. I'll happily take this. So far there have only been minor things that can easily be fixed.

author reviewer
#53 lore raw
Quoted 18 lines · Christian Brauner
Quoted 17 lines · Jan Kara

Hello,

this is a v3 of the patch series which implements the idea of blkdev_get_by_*() calls returning bdev_handle which is then passed to blkdev_put() [1]. This makes the get and put calls for bdevs more obviously matching and allows us to propagate context from get to put without having to modify all the users (again!). In particular I need to propagate used open flags to blkdev_put() to be able count writeable opens and add support for blocking writes to mounted block devices. I'll send that series separately.

The series is based on Christian's vfs tree as of today as there is quite some overlap. Patches have passed some reasonable testing - I've tested block changes, md, dm, bcache, xfs, btrfs, ext4, swap. More testing or review is always welcome. Thanks! I've pushed out the full branch to:

git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs.git bdev_handle

Quoted 8 lines
Quoted 4 lines

to ease review / testing. Since there were not many comments for v2 and Christoph has acked the series I think we should start discussing how to merge the series. Most collisions with this series seem to happen in the filesystems area so VFS tree would seem as the least painful way to merge this. Jens,

I really do like this series especially struct bdev_handle and moving the mode bits in there. I'll happily take this. So far there have only been minor things that can easily be fixed.

Thanks. Since Al is fine with just doing a potential conversion to 'struct file' as a handle on top of this series (it will be dumb Coccinelle replacement) I think we can go ahead with the series as is. As you said there will be some conflicts in btrfs and I've learned about f2fs conflicts as well so I can rebase & repost the series on top of rc1 to make life easier for you.

Honza

Signature · 2 lines
Jan Kara <jack-IBi9RG/[email protected]>
SUSE Labs, CR
#54 lore raw
Quoted 4 lines

replacement) I think we can go ahead with the series as is. As you said there will be some conflicts in btrfs and I've learned about f2fs conflicts as well so I can rebase & repost the series on top of rc1 to make life easier for you.

That is be much appreciated. Thank you!

author reviewer
#55 lore raw

PATCH v3 15/29 scsi: target: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert iblock and pscsi drivers to use bdev_open_by_path() and pass the handle around.

---
 drivers/target/target_core_iblock.c | 19 +++++++++++--------
 drivers/target/target_core_iblock.h |  1 +
 drivers/target/target_core_pscsi.c  | 26 +++++++++++++-------------
 drivers/target/target_core_pscsi.h  |  2 +-
 4 files changed, 26 insertions(+), 22 deletions(-)
drivers/target/target_core_iblock.c +11 -8
index a6a06a5f7483..8eb9eb7ce5df 100644
@@ -91,7 +91,8 @@ static int iblock_configure_device(struct se_device *dev)
 {
 	struct iblock_dev *ib_dev = IBLOCK_DEV(dev);
 	struct request_queue *q;
-	struct block_device *bd = NULL;
+	struct bdev_handle *bdev_handle;
+	struct block_device *bd;
 	struct blk_integrity *bi;
 	blk_mode_t mode = BLK_OPEN_READ;
 	unsigned int max_write_zeroes_sectors;
@@ -116,12 +117,14 @@ static int iblock_configure_device(struct se_device *dev)
 	else
 		dev->dev_flags |= DF_READ_ONLY;
 
-	bd = blkdev_get_by_path(ib_dev->ibd_udev_path, mode, ib_dev, NULL);
-	if (IS_ERR(bd)) {
-		ret = PTR_ERR(bd);
+	bdev_handle = bdev_open_by_path(ib_dev->ibd_udev_path, mode, ib_dev,
+					NULL);
+	if (IS_ERR(bdev_handle)) {
+		ret = PTR_ERR(bdev_handle);
 		goto out_free_bioset;
 	}
-	ib_dev->ibd_bd = bd;
+	ib_dev->ibd_bdev_handle = bdev_handle;
+	ib_dev->ibd_bd = bd = bdev_handle->bdev;
 
 	q = bdev_get_queue(bd);
 
@@ -177,7 +180,7 @@ static int iblock_configure_device(struct se_device *dev)
 	return 0;
 
 out_blkdev_put:
-	blkdev_put(ib_dev->ibd_bd, ib_dev);
+	bdev_release(ib_dev->ibd_bdev_handle);
 out_free_bioset:
 	bioset_exit(&ib_dev->ibd_bio_set);
 out:
@@ -202,8 +205,8 @@ static void iblock_destroy_device(struct se_device *dev)
 {
 	struct iblock_dev *ib_dev = IBLOCK_DEV(dev);
 
-	if (ib_dev->ibd_bd != NULL)
-		blkdev_put(ib_dev->ibd_bd, ib_dev);
+	if (ib_dev->ibd_bdev_handle)
+		bdev_release(ib_dev->ibd_bdev_handle);
 	bioset_exit(&ib_dev->ibd_bio_set);
 }
 
drivers/target/target_core_iblock.h +1 -0
index 8c55375d2f75..683f9a55945b 100644
@@ -32,6 +32,7 @@ struct iblock_dev {
 	u32	ibd_flags;
 	struct bio_set	ibd_bio_set;
 	struct block_device *ibd_bd;
+	struct bdev_handle *ibd_bdev_handle;
 	bool ibd_readonly;
 	struct iblock_dev_plug *ibd_plug;
 } ____cacheline_aligned;
drivers/target/target_core_pscsi.c +13 -13
index 0d4f09693ef4..41b7489d37ce 100644
@@ -352,7 +352,7 @@ static int pscsi_create_type_disk(struct se_device *dev, struct scsi_device *sd)
 	struct pscsi_hba_virt *phv = dev->se_hba->hba_ptr;
 	struct pscsi_dev_virt *pdv = PSCSI_DEV(dev);
 	struct Scsi_Host *sh = sd->host;
-	struct block_device *bd;
+	struct bdev_handle *bdev_handle;
 	int ret;
 
 	if (scsi_device_get(sd)) {
@@ -366,18 +366,18 @@ static int pscsi_create_type_disk(struct se_device *dev, struct scsi_device *sd)
 	 * Claim exclusive struct block_device access to struct scsi_device
 	 * for TYPE_DISK and TYPE_ZBC using supplied udev_path
 	 */
-	bd = blkdev_get_by_path(dev->udev_path, BLK_OPEN_WRITE | BLK_OPEN_READ,
-				pdv, NULL);
-	if (IS_ERR(bd)) {
-		pr_err("pSCSI: blkdev_get_by_path() failed\n");
+	bdev_handle = bdev_open_by_path(dev->udev_path,
+				BLK_OPEN_WRITE | BLK_OPEN_READ, pdv, NULL);
+	if (IS_ERR(bdev_handle)) {
+		pr_err("pSCSI: bdev_open_by_path() failed\n");
 		scsi_device_put(sd);
-		return PTR_ERR(bd);
+		return PTR_ERR(bdev_handle);
 	}
-	pdv->pdv_bd = bd;
+	pdv->pdv_bdev_handle = bdev_handle;
 
 	ret = pscsi_add_device_to_list(dev, sd);
 	if (ret) {
-		blkdev_put(pdv->pdv_bd, pdv);
+		bdev_release(bdev_handle);
 		scsi_device_put(sd);
 		return ret;
 	}
@@ -564,9 +564,9 @@ static void pscsi_destroy_device(struct se_device *dev)
 		 * from pscsi_create_type_disk()
 		 */
 		if ((sd->type == TYPE_DISK || sd->type == TYPE_ZBC) &&
-		    pdv->pdv_bd) {
-			blkdev_put(pdv->pdv_bd, pdv);
-			pdv->pdv_bd = NULL;
+		    pdv->pdv_bdev_handle) {
+			bdev_release(pdv->pdv_bdev_handle);
+			pdv->pdv_bdev_handle = NULL;
 		}
 		/*
 		 * For HBA mode PHV_LLD_SCSI_HOST_NO, release the reference
@@ -994,8 +994,8 @@ static sector_t pscsi_get_blocks(struct se_device *dev)
 {
 	struct pscsi_dev_virt *pdv = PSCSI_DEV(dev);
 
-	if (pdv->pdv_bd)
-		return bdev_nr_sectors(pdv->pdv_bd);
+	if (pdv->pdv_bdev_handle)
+		return bdev_nr_sectors(pdv->pdv_bdev_handle->bdev);
 	return 0;
 }
 
drivers/target/target_core_pscsi.h +1 -1
index 23d9a6e340d4..b0a3ef136592 100644
@@ -37,7 +37,7 @@ struct pscsi_dev_virt {
 	int	pdv_channel_id;
 	int	pdv_target_id;
 	int	pdv_lun_id;
-	struct block_device *pdv_bd;
+	struct bdev_handle *pdv_bdev_handle;
 	struct scsi_device *pdv_sd;
 	struct Scsi_Host *pdv_lld_host;
 } ____cacheline_aligned;
Signature · 1 line
2.35.3
Reviewed-by on v3 15/29
#56 lore raw

Re: [PATCH 15/29] scsi: target: Convert to bdev_open_by_path()

Jan,

Quoted 2 lines

Convert iblock and pscsi drivers to use bdev_open_by_path() and pass the handle around.

Looks OK to me.

Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering
author reviewer
#57 lore raw

PATCH v3 16/29 PM: hibernate: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert hibernation code to use bdev_open_by_dev().

---
 kernel/power/swap.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)
kernel/power/swap.c +16 -15
index f6ebcd00c410..b475bee282ff 100644
@@ -222,7 +222,7 @@ int swsusp_swap_in_use(void)
  */
 
 static unsigned short root_swap = 0xffff;
-static struct block_device *hib_resume_bdev;
+static struct bdev_handle *hib_resume_bdev_handle;
 
 struct hib_bio_batch {
 	atomic_t		count;
@@ -276,7 +276,8 @@ static int hib_submit_io(blk_opf_t opf, pgoff_t page_off, void *addr,
 	struct bio *bio;
 	int error = 0;
 
-	bio = bio_alloc(hib_resume_bdev, 1, opf, GFP_NOIO | __GFP_HIGH);
+	bio = bio_alloc(hib_resume_bdev_handle->bdev, 1, opf,
+			GFP_NOIO | __GFP_HIGH);
 	bio->bi_iter.bi_sector = page_off * (PAGE_SIZE >> 9);
 
 	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
@@ -356,14 +357,14 @@ static int swsusp_swap_check(void)
 		return res;
 	root_swap = res;
 
-	hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device,
+	hib_resume_bdev_handle = bdev_open_by_dev(swsusp_resume_device,
 			BLK_OPEN_WRITE, NULL, NULL);
-	if (IS_ERR(hib_resume_bdev))
-		return PTR_ERR(hib_resume_bdev);
+	if (IS_ERR(hib_resume_bdev_handle))
+		return PTR_ERR(hib_resume_bdev_handle);
 
-	res = set_blocksize(hib_resume_bdev, PAGE_SIZE);
+	res = set_blocksize(hib_resume_bdev_handle->bdev, PAGE_SIZE);
 	if (res < 0)
-		blkdev_put(hib_resume_bdev, NULL);
+		bdev_release(hib_resume_bdev_handle);
 
 	return res;
 }
@@ -1521,10 +1522,10 @@ int swsusp_check(bool snapshot_test)
 	void *holder = snapshot_test ? &swsusp_holder : NULL;
 	int error;
 
-	hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device, BLK_OPEN_READ,
-					    holder, NULL);
-	if (!IS_ERR(hib_resume_bdev)) {
-		set_blocksize(hib_resume_bdev, PAGE_SIZE);
+	hib_resume_bdev_handle = bdev_open_by_dev(swsusp_resume_device,
+				BLK_OPEN_READ, holder, NULL);
+	if (!IS_ERR(hib_resume_bdev_handle)) {
+		set_blocksize(hib_resume_bdev_handle->bdev, PAGE_SIZE);
 		clear_page(swsusp_header);
 		error = hib_submit_io(REQ_OP_READ, swsusp_resume_block,
 					swsusp_header, NULL);
@@ -1549,11 +1550,11 @@ int swsusp_check(bool snapshot_test)
 
 put:
 		if (error)
-			blkdev_put(hib_resume_bdev, holder);
+			bdev_release(hib_resume_bdev_handle);
 		else
 			pr_debug("Image signature found, resuming\n");
 	} else {
-		error = PTR_ERR(hib_resume_bdev);
+		error = PTR_ERR(hib_resume_bdev_handle);
 	}
 
 	if (error)
@@ -1568,12 +1569,12 @@ int swsusp_check(bool snapshot_test)
 
 void swsusp_close(bool snapshot_test)
 {
-	if (IS_ERR(hib_resume_bdev)) {
+	if (IS_ERR(hib_resume_bdev_handle)) {
 		pr_debug("Image device not initialised\n");
 		return;
 	}
 
-	blkdev_put(hib_resume_bdev, snapshot_test ? &swsusp_holder : NULL);
+	bdev_release(hib_resume_bdev_handle);
 }
 
 /**
Signature · 1 line
2.35.3
author reviewer
#59 lore raw

PATCH v3 16/29 PM: hibernate: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert hibernation code to use bdev_open_by_dev().

---
 kernel/power/swap.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)
kernel/power/swap.c +16 -15
index 74edbce2320b..095a263da7c4 100644
@@ -222,7 +222,7 @@ int swsusp_swap_in_use(void)
  */
 
 static unsigned short root_swap = 0xffff;
-static struct block_device *hib_resume_bdev;
+static struct bdev_handle *hib_resume_bdev_handle;
 
 struct hib_bio_batch {
 	atomic_t		count;
@@ -276,7 +276,8 @@ static int hib_submit_io(blk_opf_t opf, pgoff_t page_off, void *addr,
 	struct bio *bio;
 	int error = 0;
 
-	bio = bio_alloc(hib_resume_bdev, 1, opf, GFP_NOIO | __GFP_HIGH);
+	bio = bio_alloc(hib_resume_bdev_handle->bdev, 1, opf,
+			GFP_NOIO | __GFP_HIGH);
 	bio->bi_iter.bi_sector = page_off * (PAGE_SIZE >> 9);
 
 	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
@@ -356,14 +357,14 @@ static int swsusp_swap_check(void)
 		return res;
 	root_swap = res;
 
-	hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device,
+	hib_resume_bdev_handle = bdev_open_by_dev(swsusp_resume_device,
 			BLK_OPEN_WRITE, NULL, NULL);
-	if (IS_ERR(hib_resume_bdev))
-		return PTR_ERR(hib_resume_bdev);
+	if (IS_ERR(hib_resume_bdev_handle))
+		return PTR_ERR(hib_resume_bdev_handle);
 
-	res = set_blocksize(hib_resume_bdev, PAGE_SIZE);
+	res = set_blocksize(hib_resume_bdev_handle->bdev, PAGE_SIZE);
 	if (res < 0)
-		blkdev_put(hib_resume_bdev, NULL);
+		bdev_release(hib_resume_bdev_handle);
 
 	return res;
 }
@@ -1522,10 +1523,10 @@ int swsusp_check(bool exclusive)
 	void *holder = exclusive ? &swsusp_holder : NULL;
 	int error;
 
-	hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device, BLK_OPEN_READ,
-					    holder, NULL);
-	if (!IS_ERR(hib_resume_bdev)) {
-		set_blocksize(hib_resume_bdev, PAGE_SIZE);
+	hib_resume_bdev_handle = bdev_open_by_dev(swsusp_resume_device,
+				BLK_OPEN_READ, holder, NULL);
+	if (!IS_ERR(hib_resume_bdev_handle)) {
+		set_blocksize(hib_resume_bdev_handle->bdev, PAGE_SIZE);
 		clear_page(swsusp_header);
 		error = hib_submit_io(REQ_OP_READ, swsusp_resume_block,
 					swsusp_header, NULL);
@@ -1550,11 +1551,11 @@ int swsusp_check(bool exclusive)
 
 put:
 		if (error)
-			blkdev_put(hib_resume_bdev, holder);
+			bdev_release(hib_resume_bdev_handle);
 		else
 			pr_debug("Image signature found, resuming\n");
 	} else {
-		error = PTR_ERR(hib_resume_bdev);
+		error = PTR_ERR(hib_resume_bdev_handle);
 	}
 
 	if (error)
@@ -1570,12 +1571,12 @@ int swsusp_check(bool exclusive)
 
 void swsusp_close(bool exclusive)
 {
-	if (IS_ERR(hib_resume_bdev)) {
+	if (IS_ERR(hib_resume_bdev_handle)) {
 		pr_debug("Image device not initialised\n");
 		return;
 	}
 
-	blkdev_put(hib_resume_bdev, exclusive ? &swsusp_holder : NULL);
+	bdev_release(hib_resume_bdev_handle);
 }
 
 /**
Signature · 1 line
2.35.3
author reviewer
#60 lore raw

PATCH v3 17/29 PM: hibernate: Drop unused snapshot_test argument

In reply to Jan Kara

snapshot_test argument is now unused in swsusp_close() and load_image_and_restore(). Drop it

---
 kernel/power/hibernate.c | 14 +++++++-------
 kernel/power/power.h     |  2 +-
 kernel/power/swap.c      |  6 +++---
 3 files changed, 11 insertions(+), 11 deletions(-)
kernel/power/hibernate.c +7 -7
index e1b4bfa938dd..6abeec0ae084 100644
@@ -684,7 +684,7 @@ static void power_down(void)
 		cpu_relax();
 }
 
-static int load_image_and_restore(bool snapshot_test)
+static int load_image_and_restore(void)
 {
 	int error;
 	unsigned int flags;
@@ -694,12 +694,12 @@ static int load_image_and_restore(bool snapshot_test)
 	lock_device_hotplug();
 	error = create_basic_memory_bitmaps();
 	if (error) {
-		swsusp_close(snapshot_test);
+		swsusp_close();
 		goto Unlock;
 	}
 
 	error = swsusp_read(&flags);
-	swsusp_close(snapshot_test);
+	swsusp_close();
 	if (!error)
 		error = hibernation_restore(flags & SF_PLATFORM_MODE);
 
@@ -788,7 +788,7 @@ int hibernate(void)
 		pm_pr_dbg("Checking hibernation image\n");
 		error = swsusp_check(snapshot_test);
 		if (!error)
-			error = load_image_and_restore(snapshot_test);
+			error = load_image_and_restore();
 	}
 	thaw_processes();
 
@@ -952,7 +952,7 @@ static int software_resume(void)
 	/* The snapshot device should not be opened while we're running */
 	if (!hibernate_acquire()) {
 		error = -EBUSY;
-		swsusp_close(false);
+		swsusp_close();
 		goto Unlock;
 	}
 
@@ -973,7 +973,7 @@ static int software_resume(void)
 		goto Close_Finish;
 	}
 
-	error = load_image_and_restore(false);
+	error = load_image_and_restore();
 	thaw_processes();
  Finish:
 	pm_notifier_call_chain(PM_POST_RESTORE);
@@ -987,7 +987,7 @@ static int software_resume(void)
 	pm_pr_dbg("Hibernation image not present or could not be loaded.\n");
 	return error;
  Close_Finish:
-	swsusp_close(false);
+	swsusp_close();
 	goto Finish;
 }
 
kernel/power/power.h +1 -1
index 46eb14dc50c3..bebf049a51c1 100644
@@ -172,7 +172,7 @@ int swsusp_check(bool snapshot_test);
 extern void swsusp_free(void);
 extern int swsusp_read(unsigned int *flags_p);
 extern int swsusp_write(unsigned int flags);
-void swsusp_close(bool snapshot_test);
+void swsusp_close(void);
 #ifdef CONFIG_SUSPEND
 extern int swsusp_unmark(void);
 #endif
kernel/power/swap.c +3 -3
index b475bee282ff..17e0dad5008e 100644
@@ -444,7 +444,7 @@ static int get_swap_writer(struct swap_map_handle *handle)
 err_rel:
 	release_swap_writer(handle);
 err_close:
-	swsusp_close(false);
+	swsusp_close();
 	return ret;
 }
 
@@ -509,7 +509,7 @@ static int swap_writer_finish(struct swap_map_handle *handle,
 	if (error)
 		free_all_swap_pages(root_swap);
 	release_swap_writer(handle);
-	swsusp_close(false);
+	swsusp_close();
 
 	return error;
 }
@@ -1567,7 +1567,7 @@ int swsusp_check(bool snapshot_test)
  *	swsusp_close - close swap device.
  */
 
-void swsusp_close(bool snapshot_test)
+void swsusp_close(void)
 {
 	if (IS_ERR(hib_resume_bdev_handle)) {
 		pr_debug("Image device not initialised\n");
Signature · 1 line
2.35.3
author reviewer
#62 lore raw

PATCH v3 17/29 PM: hibernate: Drop unused snapshot_test argument

In reply to Jan Kara

snapshot_test argument is now unused in swsusp_close() and load_image_and_restore(). Drop it

---
 kernel/power/hibernate.c | 14 +++++++-------
 kernel/power/power.h     |  2 +-
 kernel/power/swap.c      |  6 +++---
 3 files changed, 11 insertions(+), 11 deletions(-)
kernel/power/hibernate.c +7 -7
index 8d35b9f9aaa3..dee341ae4ace 100644
@@ -684,7 +684,7 @@ static void power_down(void)
 		cpu_relax();
 }
 
-static int load_image_and_restore(bool snapshot_test)
+static int load_image_and_restore(void)
 {
 	int error;
 	unsigned int flags;
@@ -694,12 +694,12 @@ static int load_image_and_restore(bool snapshot_test)
 	lock_device_hotplug();
 	error = create_basic_memory_bitmaps();
 	if (error) {
-		swsusp_close(snapshot_test);
+		swsusp_close();
 		goto Unlock;
 	}
 
 	error = swsusp_read(&flags);
-	swsusp_close(snapshot_test);
+	swsusp_close();
 	if (!error)
 		error = hibernation_restore(flags & SF_PLATFORM_MODE);
 
@@ -788,7 +788,7 @@ int hibernate(void)
 		pm_pr_dbg("Checking hibernation image\n");
 		error = swsusp_check(false);
 		if (!error)
-			error = load_image_and_restore(false);
+			error = load_image_and_restore();
 	}
 	thaw_processes();
 
@@ -952,7 +952,7 @@ static int software_resume(void)
 	/* The snapshot device should not be opened while we're running */
 	if (!hibernate_acquire()) {
 		error = -EBUSY;
-		swsusp_close(true);
+		swsusp_close();
 		goto Unlock;
 	}
 
@@ -973,7 +973,7 @@ static int software_resume(void)
 		goto Close_Finish;
 	}
 
-	error = load_image_and_restore(true);
+	error = load_image_and_restore();
 	thaw_processes();
  Finish:
 	pm_notifier_call_chain(PM_POST_RESTORE);
@@ -987,7 +987,7 @@ static int software_resume(void)
 	pm_pr_dbg("Hibernation image not present or could not be loaded.\n");
 	return error;
  Close_Finish:
-	swsusp_close(true);
+	swsusp_close();
 	goto Finish;
 }
 
kernel/power/power.h +1 -1
index a98f95e309a3..17fd9aaaf084 100644
@@ -172,7 +172,7 @@ int swsusp_check(bool exclusive);
 extern void swsusp_free(void);
 extern int swsusp_read(unsigned int *flags_p);
 extern int swsusp_write(unsigned int flags);
-void swsusp_close(bool exclusive);
+void swsusp_close(void);
 #ifdef CONFIG_SUSPEND
 extern int swsusp_unmark(void);
 #endif
kernel/power/swap.c +3 -3
index 095a263da7c4..68a5c2f06957 100644
@@ -444,7 +444,7 @@ static int get_swap_writer(struct swap_map_handle *handle)
 err_rel:
 	release_swap_writer(handle);
 err_close:
-	swsusp_close(false);
+	swsusp_close();
 	return ret;
 }
 
@@ -509,7 +509,7 @@ static int swap_writer_finish(struct swap_map_handle *handle,
 	if (error)
 		free_all_swap_pages(root_swap);
 	release_swap_writer(handle);
-	swsusp_close(false);
+	swsusp_close();
 
 	return error;
 }
@@ -1569,7 +1569,7 @@ int swsusp_check(bool exclusive)
  * @exclusive: Close the resume device which is exclusively opened.
  */
 
-void swsusp_close(bool exclusive)
+void swsusp_close(void)
 {
 	if (IS_ERR(hib_resume_bdev_handle)) {
 		pr_debug("Image device not initialised\n");
Signature · 1 line
2.35.3
author reviewer
#63 lore raw

PATCH v3 18/29 mm/swap: Convert to use bdev_open_by_dev()

In reply to Jan Kara

Convert swapping code to use bdev_open_by_dev() and pass the handle around.

---
 include/linux/swap.h |  1 +
 mm/swapfile.c        | 23 ++++++++++++-----------
 2 files changed, 13 insertions(+), 11 deletions(-)
include/linux/swap.h +1 -0
index 456546443f1f..62334f8d4932 100644
@@ -298,6 +298,7 @@ struct swap_info_struct {
 	unsigned int __percpu *cluster_next_cpu; /*percpu index for next allocation */
 	struct percpu_cluster __percpu *percpu_cluster; /* per cpu's swap location */
 	struct rb_root swap_extent_root;/* root of the swap extent rbtree */
+	struct bdev_handle *bdev_handle;/* open handle of the bdev */
 	struct block_device *bdev;	/* swap device or bdev of swap file */
 	struct file *swap_file;		/* seldom referenced */
 	unsigned int old_block_size;	/* seldom referenced */
mm/swapfile.c +12 -11
index 8e6dde68b389..621428ee6895 100644
@@ -2526,11 +2526,10 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 	exit_swap_address_space(p->type);
 
 	inode = mapping->host;
-	if (S_ISBLK(inode->i_mode)) {
-		struct block_device *bdev = I_BDEV(inode);
-
-		set_blocksize(bdev, old_block_size);
-		blkdev_put(bdev, p);
+	if (p->bdev_handle) {
+		set_blocksize(p->bdev, old_block_size);
+		bdev_release(p->bdev_handle);
+		p->bdev_handle = NULL;
 	}
 
 	inode_lock(inode);
@@ -2760,13 +2759,14 @@ static int claim_swapfile(struct swap_info_struct *p, struct inode *inode)
 	int error;
 
 	if (S_ISBLK(inode->i_mode)) {
-		p->bdev = blkdev_get_by_dev(inode->i_rdev,
+		p->bdev_handle = bdev_open_by_dev(inode->i_rdev,
 				BLK_OPEN_READ | BLK_OPEN_WRITE, p, NULL);
-		if (IS_ERR(p->bdev)) {
-			error = PTR_ERR(p->bdev);
-			p->bdev = NULL;
+		if (IS_ERR(p->bdev_handle)) {
+			error = PTR_ERR(p->bdev_handle);
+			p->bdev_handle = NULL;
 			return error;
 		}
+		p->bdev = p->bdev_handle->bdev;
 		p->old_block_size = block_size(p->bdev);
 		error = set_blocksize(p->bdev, PAGE_SIZE);
 		if (error < 0)
@@ -3210,9 +3210,10 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
 	p->percpu_cluster = NULL;
 	free_percpu(p->cluster_next_cpu);
 	p->cluster_next_cpu = NULL;
-	if (inode && S_ISBLK(inode->i_mode) && p->bdev) {
+	if (p->bdev_handle) {
 		set_blocksize(p->bdev, p->old_block_size);
-		blkdev_put(p->bdev, p);
+		bdev_release(p->bdev_handle);
+		p->bdev_handle = NULL;
 	}
 	inode = NULL;
 	destroy_swap_extents(p);
Signature · 1 line
2.35.3
author reviewer
#65 lore raw

PATCH v3 18/29 mm/swap: Convert to use bdev_open_by_dev()

In reply to Jan Kara

Convert swapping code to use bdev_open_by_dev() and pass the handle around.

---
 include/linux/swap.h |  1 +
 mm/swapfile.c        | 23 ++++++++++++-----------
 2 files changed, 13 insertions(+), 11 deletions(-)
include/linux/swap.h +1 -0
index 493487ed7c38..f6dd6575b905 100644
@@ -298,6 +298,7 @@ struct swap_info_struct {
 	unsigned int __percpu *cluster_next_cpu; /*percpu index for next allocation */
 	struct percpu_cluster __percpu *percpu_cluster; /* per cpu's swap location */
 	struct rb_root swap_extent_root;/* root of the swap extent rbtree */
+	struct bdev_handle *bdev_handle;/* open handle of the bdev */
 	struct block_device *bdev;	/* swap device or bdev of swap file */
 	struct file *swap_file;		/* seldom referenced */
 	unsigned int old_block_size;	/* seldom referenced */
mm/swapfile.c +12 -11
index e52f486834eb..4bc70f459164 100644
@@ -2530,11 +2530,10 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 	exit_swap_address_space(p->type);
 
 	inode = mapping->host;
-	if (S_ISBLK(inode->i_mode)) {
-		struct block_device *bdev = I_BDEV(inode);
-
-		set_blocksize(bdev, old_block_size);
-		blkdev_put(bdev, p);
+	if (p->bdev_handle) {
+		set_blocksize(p->bdev, old_block_size);
+		bdev_release(p->bdev_handle);
+		p->bdev_handle = NULL;
 	}
 
 	inode_lock(inode);
@@ -2764,13 +2763,14 @@ static int claim_swapfile(struct swap_info_struct *p, struct inode *inode)
 	int error;
 
 	if (S_ISBLK(inode->i_mode)) {
-		p->bdev = blkdev_get_by_dev(inode->i_rdev,
+		p->bdev_handle = bdev_open_by_dev(inode->i_rdev,
 				BLK_OPEN_READ | BLK_OPEN_WRITE, p, NULL);
-		if (IS_ERR(p->bdev)) {
-			error = PTR_ERR(p->bdev);
-			p->bdev = NULL;
+		if (IS_ERR(p->bdev_handle)) {
+			error = PTR_ERR(p->bdev_handle);
+			p->bdev_handle = NULL;
 			return error;
 		}
+		p->bdev = p->bdev_handle->bdev;
 		p->old_block_size = block_size(p->bdev);
 		error = set_blocksize(p->bdev, PAGE_SIZE);
 		if (error < 0)
@@ -3206,9 +3206,10 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
 	p->percpu_cluster = NULL;
 	free_percpu(p->cluster_next_cpu);
 	p->cluster_next_cpu = NULL;
-	if (inode && S_ISBLK(inode->i_mode) && p->bdev) {
+	if (p->bdev_handle) {
 		set_blocksize(p->bdev, p->old_block_size);
-		blkdev_put(p->bdev, p);
+		bdev_release(p->bdev_handle);
+		p->bdev_handle = NULL;
 	}
 	inode = NULL;
 	destroy_swap_extents(p);
Signature · 1 line
2.35.3
author reviewer
#66 lore raw

PATCH v3 19/29 fs: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert mount code to use bdev_open_by_dev() and propagate the handle around to bdev_release().

---
 fs/cramfs/inode.c  |  2 +-
 fs/romfs/super.c   |  2 +-
 fs/super.c         | 15 +++++++++------
 include/linux/fs.h |  1 +
 4 files changed, 12 insertions(+), 8 deletions(-)
fs/cramfs/inode.c +1 -1
index 5ee7d7bbb361..2fbf97077ce9 100644
@@ -495,7 +495,7 @@ static void cramfs_kill_sb(struct super_block *sb)
 		sb->s_mtd = NULL;
 	} else if (IS_ENABLED(CONFIG_CRAMFS_BLOCKDEV) && sb->s_bdev) {
 		sync_blockdev(sb->s_bdev);
-		blkdev_put(sb->s_bdev, sb);
+		bdev_release(sb->s_bdev_handle);
 	}
 	kfree(sbi);
 }
fs/romfs/super.c +1 -1
index 5c35f6c76037..b1bdfbc211c3 100644
@@ -593,7 +593,7 @@ static void romfs_kill_sb(struct super_block *sb)
 #ifdef CONFIG_ROMFS_ON_BLOCK
 	if (sb->s_bdev) {
 		sync_blockdev(sb->s_bdev);
-		blkdev_put(sb->s_bdev, sb);
+		bdev_release(sb->s_bdev_handle);
 	}
 #endif
 }
fs/super.c +9 -6
index 1bea34be5d43..82757f3e0f1b 100644
@@ -1250,14 +1250,16 @@ int setup_bdev_super(struct super_block *sb, int sb_flags,
 		struct fs_context *fc)
 {
 	blk_mode_t mode = sb_open_mode(sb_flags);
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 
-	bdev = blkdev_get_by_dev(sb->s_dev, mode, sb, &fs_holder_ops);
-	if (IS_ERR(bdev)) {
+	bdev_handle = bdev_open_by_dev(sb->s_dev, mode, sb, &fs_holder_ops);
+	if (IS_ERR(bdev_handle)) {
 		if (fc)
 			errorf(fc, "%s: Can't open blockdev", fc->source);
-		return PTR_ERR(bdev);
+		return PTR_ERR(bdev_handle);
 	}
+	bdev = bdev_handle->bdev;
 
 	/*
 	 * This really should be in blkdev_get_by_dev, but right now can't due
@@ -1265,7 +1267,7 @@ int setup_bdev_super(struct super_block *sb, int sb_flags,
 	 * writable from userspace even for a read-only block device.
 	 */
 	if ((mode & BLK_OPEN_WRITE) && bdev_read_only(bdev)) {
-		blkdev_put(bdev, sb);
+		bdev_release(bdev_handle);
 		return -EACCES;
 	}
 
@@ -1281,10 +1283,11 @@ int setup_bdev_super(struct super_block *sb, int sb_flags,
 		mutex_unlock(&bdev->bd_fsfreeze_mutex);
 		if (fc)
 			warnf(fc, "%pg: Can't mount, blockdev is frozen", bdev);
-		blkdev_put(bdev, sb);
+		bdev_release(bdev_handle);
 		return -EBUSY;
 	}
 	spin_lock(&sb_lock);
+	sb->s_bdev_handle = bdev_handle;
 	sb->s_bdev = bdev;
 	sb->s_bdi = bdi_get(bdev->bd_disk->bdi);
 	if (bdev_stable_writes(bdev))
@@ -1418,7 +1421,7 @@ void kill_block_super(struct super_block *sb)
 	generic_shutdown_super(sb);
 	if (bdev) {
 		sync_blockdev(bdev);
-		blkdev_put(bdev, sb);
+		bdev_release(sb->s_bdev_handle);
 	}
 }
 
include/linux/fs.h +1 -0
index 22b59cd36963..b5c61107bd6d 100644
@@ -1187,6 +1187,7 @@ struct super_block {
 	struct hlist_bl_head	s_roots;	/* alternate root dentries for NFS */
 	struct list_head	s_mounts;	/* list of mounts; _not_ for fs use */
 	struct block_device	*s_bdev;
+	struct bdev_handle	*s_bdev_handle;
 	struct backing_dev_info *s_bdi;
 	struct mtd_info		*s_mtd;
 	struct hlist_node	s_instances;
Signature · 1 line
2.35.3
author reviewer
#68 lore raw

PATCH v3 19/29 fs: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert mount code to use bdev_open_by_dev() and propagate the handle around to bdev_release().

---
 fs/cramfs/inode.c  |  2 +-
 fs/romfs/super.c   |  2 +-
 fs/super.c         | 15 +++++++++------
 include/linux/fs.h |  1 +
 4 files changed, 12 insertions(+), 8 deletions(-)
fs/cramfs/inode.c +1 -1
index 5ee7d7bbb361..2fbf97077ce9 100644
@@ -495,7 +495,7 @@ static void cramfs_kill_sb(struct super_block *sb)
 		sb->s_mtd = NULL;
 	} else if (IS_ENABLED(CONFIG_CRAMFS_BLOCKDEV) && sb->s_bdev) {
 		sync_blockdev(sb->s_bdev);
-		blkdev_put(sb->s_bdev, sb);
+		bdev_release(sb->s_bdev_handle);
 	}
 	kfree(sbi);
 }
fs/romfs/super.c +1 -1
index 5c35f6c76037..b1bdfbc211c3 100644
@@ -593,7 +593,7 @@ static void romfs_kill_sb(struct super_block *sb)
 #ifdef CONFIG_ROMFS_ON_BLOCK
 	if (sb->s_bdev) {
 		sync_blockdev(sb->s_bdev);
-		blkdev_put(sb->s_bdev, sb);
+		bdev_release(sb->s_bdev_handle);
 	}
 #endif
 }
fs/super.c +9 -6
index 2d762ce67f6e..26b96191e9b3 100644
@@ -1479,14 +1479,16 @@ int setup_bdev_super(struct super_block *sb, int sb_flags,
 		struct fs_context *fc)
 {
 	blk_mode_t mode = sb_open_mode(sb_flags);
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 
-	bdev = blkdev_get_by_dev(sb->s_dev, mode, sb, &fs_holder_ops);
-	if (IS_ERR(bdev)) {
+	bdev_handle = bdev_open_by_dev(sb->s_dev, mode, sb, &fs_holder_ops);
+	if (IS_ERR(bdev_handle)) {
 		if (fc)
 			errorf(fc, "%s: Can't open blockdev", fc->source);
-		return PTR_ERR(bdev);
+		return PTR_ERR(bdev_handle);
 	}
+	bdev = bdev_handle->bdev;
 
 	/*
 	 * This really should be in blkdev_get_by_dev, but right now can't due
@@ -1494,7 +1496,7 @@ int setup_bdev_super(struct super_block *sb, int sb_flags,
 	 * writable from userspace even for a read-only block device.
 	 */
 	if ((mode & BLK_OPEN_WRITE) && bdev_read_only(bdev)) {
-		blkdev_put(bdev, sb);
+		bdev_release(bdev_handle);
 		return -EACCES;
 	}
 
@@ -1510,10 +1512,11 @@ int setup_bdev_super(struct super_block *sb, int sb_flags,
 		mutex_unlock(&bdev->bd_fsfreeze_mutex);
 		if (fc)
 			warnf(fc, "%pg: Can't mount, blockdev is frozen", bdev);
-		blkdev_put(bdev, sb);
+		bdev_release(bdev_handle);
 		return -EBUSY;
 	}
 	spin_lock(&sb_lock);
+	sb->s_bdev_handle = bdev_handle;
 	sb->s_bdev = bdev;
 	sb->s_bdi = bdi_get(bdev->bd_disk->bdi);
 	if (bdev_stable_writes(bdev))
@@ -1646,7 +1649,7 @@ void kill_block_super(struct super_block *sb)
 	generic_shutdown_super(sb);
 	if (bdev) {
 		sync_blockdev(bdev);
-		blkdev_put(bdev, sb);
+		bdev_release(sb->s_bdev_handle);
 	}
 }
 
include/linux/fs.h +1 -0
index b528f063e8ff..03afc8b6f2af 100644
@@ -1221,6 +1221,7 @@ struct super_block {
 	struct hlist_bl_head	s_roots;	/* alternate root dentries for NFS */
 	struct list_head	s_mounts;	/* list of mounts; _not_ for fs use */
 	struct block_device	*s_bdev;
+	struct bdev_handle	*s_bdev_handle;
 	struct backing_dev_info *s_bdi;
 	struct mtd_info		*s_mtd;
 	struct hlist_node	s_instances;
Signature · 1 line
2.35.3
author reviewer
#69 lore raw

PATCH v3 20/29 btrfs: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert btrfs to use bdev_open_by_path() and pass the handle around. We also drop the holder from struct btrfs_device as it is now not needed anymore.

---
 fs/btrfs/dev-replace.c |  14 +++---
 fs/btrfs/ioctl.c       |  18 +++----
 fs/btrfs/volumes.c     | 107 +++++++++++++++++++++--------------------
 fs/btrfs/volumes.h     |   6 +--
 4 files changed, 73 insertions(+), 72 deletions(-)
fs/btrfs/dev-replace.c +8 -6
index 5f10965fd72b..fec013c5f26c 100644
@@ -247,6 +247,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
 	struct btrfs_device *device;
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 	u64 devid = BTRFS_DEV_REPLACE_DEVID;
 	int ret = 0;
@@ -257,12 +258,13 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 		return -EINVAL;
 	}
 
-	bdev = blkdev_get_by_path(device_path, BLK_OPEN_WRITE,
-				  fs_info->bdev_holder, NULL);
-	if (IS_ERR(bdev)) {
+	bdev_handle = bdev_open_by_path(device_path, BLK_OPEN_WRITE,
+					fs_info->bdev_holder, NULL);
+	if (IS_ERR(bdev_handle)) {
 		btrfs_err(fs_info, "target device %s is invalid!", device_path);
-		return PTR_ERR(bdev);
+		return PTR_ERR(bdev_handle);
 	}
+	bdev = bdev_handle->bdev;
 
 	if (!btrfs_check_device_zone_type(fs_info, bdev)) {
 		btrfs_err(fs_info,
@@ -313,9 +315,9 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	device->commit_bytes_used = device->bytes_used;
 	device->fs_info = fs_info;
 	device->bdev = bdev;
+	device->bdev_handle = bdev_handle;
 	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
 	set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
-	device->holder = fs_info->bdev_holder;
 	device->dev_stats_valid = 1;
 	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
 	device->fs_devices = fs_devices;
@@ -334,7 +336,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	return 0;
 
 error:
-	blkdev_put(bdev, fs_info->bdev_holder);
+	bdev_release(bdev_handle);
 	return ret;
 }
 
fs/btrfs/ioctl.c +8 -10
index a18ee7b5a166..b4074191fcc7 100644
@@ -2670,8 +2670,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ioctl_vol_args_v2 *vol_args;
-	struct block_device *bdev = NULL;
-	void *holder;
+	struct bdev_handle *bdev_handle = NULL;
 	int ret;
 	bool cancel = false;
 
@@ -2708,7 +2707,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 		goto err_drop;
 
 	/* Exclusive operation is now claimed */
-	ret = btrfs_rm_device(fs_info, &args, &bdev, &holder);
+	ret = btrfs_rm_device(fs_info, &args, &bdev_handle);
 
 	btrfs_exclop_finish(fs_info);
 
@@ -2722,8 +2721,8 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 	}
 err_drop:
 	mnt_drop_write_file(file);
-	if (bdev)
-		blkdev_put(bdev, holder);
+	if (bdev_handle)
+		bdev_release(bdev_handle);
 out:
 	btrfs_put_dev_args_from_path(&args);
 	kfree(vol_args);
@@ -2736,8 +2735,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ioctl_vol_args *vol_args;
-	struct block_device *bdev = NULL;
-	void *holder;
+	struct bdev_handle *bdev_handle = NULL;
 	int ret;
 	bool cancel = false;
 
@@ -2764,15 +2762,15 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	ret = exclop_start_or_cancel_reloc(fs_info, BTRFS_EXCLOP_DEV_REMOVE,
 					   cancel);
 	if (ret == 0) {
-		ret = btrfs_rm_device(fs_info, &args, &bdev, &holder);
+		ret = btrfs_rm_device(fs_info, &args, &bdev_handle);
 		if (!ret)
 			btrfs_info(fs_info, "disk deleted %s", vol_args->name);
 		btrfs_exclop_finish(fs_info);
 	}
 
 	mnt_drop_write_file(file);
-	if (bdev)
-		blkdev_put(bdev, holder);
+	if (bdev_handle)
+		bdev_release(bdev_handle);
 out:
 	btrfs_put_dev_args_from_path(&args);
 	kfree(vol_args);
fs/btrfs/volumes.c +55 -52

297 lines not on this page · raw message

fs/btrfs/volumes.h +2 -4
index f472d646715e..d92ebeaf1200 100644
@@ -90,13 +90,11 @@ struct btrfs_device {
 
 	u64 generation;
 
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 
 	struct btrfs_zoned_device_info *zone_info;
 
-	/* block device holder for blkdev_get/put */
-	void *holder;
-
 	/*
 	 * Device's major-minor number. Must be set even if the device is not
 	 * opened (bdev == NULL), unless the device is missing.
@@ -631,7 +629,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 void btrfs_put_dev_args_from_path(struct btrfs_dev_lookup_args *args);
 int btrfs_rm_device(struct btrfs_fs_info *fs_info,
 		    struct btrfs_dev_lookup_args *args,
-		    struct block_device **bdev, void **holder);
+		    struct bdev_handle **bdev_handle);
 void __exit btrfs_cleanup_fs_uuids(void);
 int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
 int btrfs_grow_device(struct btrfs_trans_handle *trans,
Signature · 1 line
2.35.3
maintainer Reviewed-by on v3 20/29
#70 lore raw

Re: [PATCH 20/29] btrfs: Convert to bdev_open_by_path()

Quoted 9 lines · Jan Kara

Convert btrfs to use bdev_open_by_path() and pass the handle around. We also drop the holder from struct btrfs_device as it is now not needed anymore.

---

Looks good to me,

fs/btrfs/volumes.c struct btrfs_device *btrfs_scan_one_device(const char *path) · line 1390

Quoted 6 lines
fs/btrfs/volumes.c +3 -3
-	bdev = blkdev_get_by_path(path, BLK_OPEN_READ, NULL, NULL);
-	if (IS_ERR(bdev))
-		return ERR_CAST(bdev);
+	bdev_handle = bdev_open_by_path(path, BLK_OPEN_READ, NULL, NULL);
+	if (IS_ERR(bdev_handle))
+		return ERR_CAST(bdev_handle);

Minor merge conflict with what's in vfs.super because we dropped unrelated btrfs patches like David asked us to. Very easy to fix for me. Just an fyi.

author reviewer
#71 lore raw

PATCH v3 20/29 btrfs: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert btrfs to use bdev_open_by_path() and pass the handle around. We also drop the holder from struct btrfs_device as it is now not needed anymore.

---
 fs/btrfs/dev-replace.c |  14 +++---
 fs/btrfs/ioctl.c       |  18 +++----
 fs/btrfs/volumes.c     | 107 +++++++++++++++++++++--------------------
 fs/btrfs/volumes.h     |   6 +--
 4 files changed, 73 insertions(+), 72 deletions(-)
fs/btrfs/dev-replace.c +8 -6
index d3998cad62c2..f9544fda38e9 100644
@@ -246,6 +246,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
 	struct btrfs_device *device;
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 	u64 devid = BTRFS_DEV_REPLACE_DEVID;
 	int ret = 0;
@@ -256,12 +257,13 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 		return -EINVAL;
 	}
 
-	bdev = blkdev_get_by_path(device_path, BLK_OPEN_WRITE,
-				  fs_info->bdev_holder, NULL);
-	if (IS_ERR(bdev)) {
+	bdev_handle = bdev_open_by_path(device_path, BLK_OPEN_WRITE,
+					fs_info->bdev_holder, NULL);
+	if (IS_ERR(bdev_handle)) {
 		btrfs_err(fs_info, "target device %s is invalid!", device_path);
-		return PTR_ERR(bdev);
+		return PTR_ERR(bdev_handle);
 	}
+	bdev = bdev_handle->bdev;
 
 	if (!btrfs_check_device_zone_type(fs_info, bdev)) {
 		btrfs_err(fs_info,
@@ -312,9 +314,9 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	device->commit_bytes_used = device->bytes_used;
 	device->fs_info = fs_info;
 	device->bdev = bdev;
+	device->bdev_handle = bdev_handle;
 	set_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state);
 	set_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state);
-	device->holder = fs_info->bdev_holder;
 	device->dev_stats_valid = 1;
 	set_blocksize(device->bdev, BTRFS_BDEV_BLOCKSIZE);
 	device->fs_devices = fs_devices;
@@ -333,7 +335,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,
 	return 0;
 
 error:
-	blkdev_put(bdev, fs_info->bdev_holder);
+	bdev_release(bdev_handle);
 	return ret;
 }
 
fs/btrfs/ioctl.c +8 -10
index 9c05216de36a..59edbf47dbb2 100644
@@ -2676,8 +2676,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ioctl_vol_args_v2 *vol_args;
-	struct block_device *bdev = NULL;
-	void *holder;
+	struct bdev_handle *bdev_handle = NULL;
 	int ret;
 	bool cancel = false;
 
@@ -2720,7 +2719,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 		goto err_drop;
 
 	/* Exclusive operation is now claimed */
-	ret = btrfs_rm_device(fs_info, &args, &bdev, &holder);
+	ret = btrfs_rm_device(fs_info, &args, &bdev_handle);
 
 	btrfs_exclop_finish(fs_info);
 
@@ -2734,8 +2733,8 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 	}
 err_drop:
 	mnt_drop_write_file(file);
-	if (bdev)
-		blkdev_put(bdev, holder);
+	if (bdev_handle)
+		bdev_release(bdev_handle);
 out:
 	btrfs_put_dev_args_from_path(&args);
 	kfree(vol_args);
@@ -2748,8 +2747,7 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	struct inode *inode = file_inode(file);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_ioctl_vol_args *vol_args;
-	struct block_device *bdev = NULL;
-	void *holder;
+	struct bdev_handle *bdev_handle = NULL;
 	int ret;
 	bool cancel = false;
 
@@ -2782,15 +2780,15 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	ret = exclop_start_or_cancel_reloc(fs_info, BTRFS_EXCLOP_DEV_REMOVE,
 					   cancel);
 	if (ret == 0) {
-		ret = btrfs_rm_device(fs_info, &args, &bdev, &holder);
+		ret = btrfs_rm_device(fs_info, &args, &bdev_handle);
 		if (!ret)
 			btrfs_info(fs_info, "disk deleted %s", vol_args->name);
 		btrfs_exclop_finish(fs_info);
 	}
 
 	mnt_drop_write_file(file);
-	if (bdev)
-		blkdev_put(bdev, holder);
+	if (bdev_handle)
+		bdev_release(bdev_handle);
 out:
 	btrfs_put_dev_args_from_path(&args);
 	kfree(vol_args);
fs/btrfs/volumes.c +55 -52

297 lines not on this page · raw message

fs/btrfs/volumes.h +2 -4
index b5d0586c82af..cb42376ad7b5 100644
@@ -90,13 +90,11 @@ struct btrfs_device {
 
 	u64 generation;
 
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 
 	struct btrfs_zoned_device_info *zone_info;
 
-	/* block device holder for blkdev_get/put */
-	void *holder;
-
 	/*
 	 * Device's major-minor number. Must be set even if the device is not
 	 * opened (bdev == NULL), unless the device is missing.
@@ -649,7 +647,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 void btrfs_put_dev_args_from_path(struct btrfs_dev_lookup_args *args);
 int btrfs_rm_device(struct btrfs_fs_info *fs_info,
 		    struct btrfs_dev_lookup_args *args,
-		    struct block_device **bdev, void **holder);
+		    struct bdev_handle **bdev_handle);
 void __exit btrfs_cleanup_fs_uuids(void);
 int btrfs_num_copies(struct btrfs_fs_info *fs_info, u64 logical, u64 len);
 int btrfs_grow_device(struct btrfs_trans_handle *trans,
Signature · 1 line
2.35.3
author reviewer
#72 lore raw

PATCH v3 21/29 erofs: Convert to use bdev_open_by_path()

In reply to Jan Kara

Convert erofs to use bdev_open_by_path() and pass the handle around.

---
 fs/erofs/data.c     |  4 ++--
 fs/erofs/internal.h |  2 +-
 fs/erofs/super.c    | 20 ++++++++++----------
 3 files changed, 13 insertions(+), 13 deletions(-)
fs/erofs/data.c +2 -2
index db5e4b7636ec..1fa60cfff267 100644
@@ -222,7 +222,7 @@ int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *map)
 			up_read(&devs->rwsem);
 			return 0;
 		}
-		map->m_bdev = dif->bdev;
+		map->m_bdev = dif->bdev_handle->bdev;
 		map->m_daxdev = dif->dax_dev;
 		map->m_dax_part_off = dif->dax_part_off;
 		map->m_fscache = dif->fscache;
@@ -240,7 +240,7 @@ int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *map)
 			if (map->m_pa >= startoff &&
 			    map->m_pa < startoff + length) {
 				map->m_pa -= startoff;
-				map->m_bdev = dif->bdev;
+				map->m_bdev = dif->bdev_handle->bdev;
 				map->m_daxdev = dif->dax_dev;
 				map->m_dax_part_off = dif->dax_part_off;
 				map->m_fscache = dif->fscache;
fs/erofs/internal.h +1 -1
index 36e32fa542f0..fabd3bb0c194 100644
@@ -47,7 +47,7 @@ typedef u32 erofs_blk_t;
 struct erofs_device_info {
 	char *path;
 	struct erofs_fscache *fscache;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct dax_device *dax_dev;
 	u64 dax_part_off;
 
fs/erofs/super.c +10 -10
index 566f68ddfa36..366a46fe7fdc 100644
@@ -230,7 +230,7 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb,
 	struct erofs_sb_info *sbi = EROFS_SB(sb);
 	struct erofs_fscache *fscache;
 	struct erofs_deviceslot *dis;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	void *ptr;
 
 	ptr = erofs_read_metabuf(buf, sb, erofs_blknr(sb, *pos), EROFS_KMAP);
@@ -254,13 +254,13 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb,
 			return PTR_ERR(fscache);
 		dif->fscache = fscache;
 	} else if (!sbi->devs->flatdev) {
-		bdev = blkdev_get_by_path(dif->path, BLK_OPEN_READ, sb->s_type,
-					  NULL);
-		if (IS_ERR(bdev))
-			return PTR_ERR(bdev);
-		dif->bdev = bdev;
-		dif->dax_dev = fs_dax_get_by_bdev(bdev, &dif->dax_part_off,
-						  NULL, NULL);
+		bdev_handle = bdev_open_by_path(dif->path, BLK_OPEN_READ,
+						sb->s_type, NULL);
+		if (IS_ERR(bdev_handle))
+			return PTR_ERR(bdev_handle);
+		dif->bdev_handle = bdev_handle;
+		dif->dax_dev = fs_dax_get_by_bdev(bdev_handle->bdev,
+				&dif->dax_part_off, NULL, NULL);
 	}
 
 	dif->blocks = le32_to_cpu(dis->blocks);
@@ -815,8 +815,8 @@ static int erofs_release_device_info(int id, void *ptr, void *data)
 	struct erofs_device_info *dif = ptr;
 
 	fs_put_dax(dif->dax_dev, NULL);
-	if (dif->bdev)
-		blkdev_put(dif->bdev, &erofs_fs_type);
+	if (dif->bdev_handle)
+		bdev_release(dif->bdev_handle);
 	erofs_fscache_unregister_cookie(dif->fscache);
 	dif->fscache = NULL;
 	kfree(dif->path);
Signature · 1 line
2.35.3
maintainer Reviewed-by on v3 21/29
#73 lore raw

Re: [PATCH 21/29] erofs: Convert to use bdev_open_by_path()

Quoted 9 lines · Jan Kara

Convert erofs to use bdev_open_by_path() and pass the handle around.

---

Looks good to me,

author reviewer
#74 lore raw

PATCH v3 21/29 erofs: Convert to use bdev_open_by_path()

In reply to Jan Kara

Convert erofs to use bdev_open_by_path() and pass the handle around.

---
 fs/erofs/data.c     |  4 ++--
 fs/erofs/internal.h |  2 +-
 fs/erofs/super.c    | 20 ++++++++++----------
 3 files changed, 13 insertions(+), 13 deletions(-)
fs/erofs/data.c +2 -2
index 0c2c99c58b5e..f6a0a1748521 100644
@@ -222,7 +222,7 @@ int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *map)
 			up_read(&devs->rwsem);
 			return 0;
 		}
-		map->m_bdev = dif->bdev;
+		map->m_bdev = dif->bdev_handle->bdev;
 		map->m_daxdev = dif->dax_dev;
 		map->m_dax_part_off = dif->dax_part_off;
 		map->m_fscache = dif->fscache;
@@ -240,7 +240,7 @@ int erofs_map_dev(struct super_block *sb, struct erofs_map_dev *map)
 			if (map->m_pa >= startoff &&
 			    map->m_pa < startoff + length) {
 				map->m_pa -= startoff;
-				map->m_bdev = dif->bdev;
+				map->m_bdev = dif->bdev_handle->bdev;
 				map->m_daxdev = dif->dax_dev;
 				map->m_dax_part_off = dif->dax_part_off;
 				map->m_fscache = dif->fscache;
fs/erofs/internal.h +1 -1
index 4ff88d0dd980..cf04e21bfda2 100644
@@ -47,7 +47,7 @@ typedef u32 erofs_blk_t;
 struct erofs_device_info {
 	char *path;
 	struct erofs_fscache *fscache;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct dax_device *dax_dev;
 	u64 dax_part_off;
 
fs/erofs/super.c +10 -10
index 44a24d573f1f..2bf6dac83025 100644
@@ -227,7 +227,7 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb,
 	struct erofs_sb_info *sbi = EROFS_SB(sb);
 	struct erofs_fscache *fscache;
 	struct erofs_deviceslot *dis;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	void *ptr;
 
 	ptr = erofs_read_metabuf(buf, sb, erofs_blknr(sb, *pos), EROFS_KMAP);
@@ -251,13 +251,13 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb,
 			return PTR_ERR(fscache);
 		dif->fscache = fscache;
 	} else if (!sbi->devs->flatdev) {
-		bdev = blkdev_get_by_path(dif->path, BLK_OPEN_READ, sb->s_type,
-					  NULL);
-		if (IS_ERR(bdev))
-			return PTR_ERR(bdev);
-		dif->bdev = bdev;
-		dif->dax_dev = fs_dax_get_by_bdev(bdev, &dif->dax_part_off,
-						  NULL, NULL);
+		bdev_handle = bdev_open_by_path(dif->path, BLK_OPEN_READ,
+						sb->s_type, NULL);
+		if (IS_ERR(bdev_handle))
+			return PTR_ERR(bdev_handle);
+		dif->bdev_handle = bdev_handle;
+		dif->dax_dev = fs_dax_get_by_bdev(bdev_handle->bdev,
+				&dif->dax_part_off, NULL, NULL);
 	}
 
 	dif->blocks = le32_to_cpu(dis->blocks);
@@ -806,8 +806,8 @@ static int erofs_release_device_info(int id, void *ptr, void *data)
 	struct erofs_device_info *dif = ptr;
 
 	fs_put_dax(dif->dax_dev, NULL);
-	if (dif->bdev)
-		blkdev_put(dif->bdev, &erofs_fs_type);
+	if (dif->bdev_handle)
+		bdev_release(dif->bdev_handle);
 	erofs_fscache_unregister_cookie(dif->fscache);
 	dif->fscache = NULL;
 	kfree(dif->path);
Signature · 1 line
2.35.3
author reviewer
#75 lore raw

PATCH v3 22/29 ext4: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert ext4 to use bdev_open_by_dev() and pass the handle around.

---
 fs/ext4/ext4.h  |  2 +-
 fs/ext4/fsmap.c |  9 +++++----
 fs/ext4/super.c | 44 +++++++++++++++++++++++---------------------
 3 files changed, 29 insertions(+), 26 deletions(-)
fs/ext4/ext4.h +1 -1
index 1e2259d9967d..6856e80abf70 100644
@@ -1541,7 +1541,7 @@ struct ext4_sb_info {
 	unsigned long s_commit_interval;
 	u32 s_max_batch_time;
 	u32 s_min_batch_time;
-	struct block_device *s_journal_bdev;
+	struct bdev_handle *s_journal_bdev_handle;
 #ifdef CONFIG_QUOTA
 	/* Names of quota files with journalled quota */
 	char __rcu *s_qf_names[EXT4_MAXQUOTAS];
fs/ext4/fsmap.c +5 -4
index cdf9bfe10137..11e6f33677a2 100644
@@ -576,8 +576,9 @@ static bool ext4_getfsmap_is_valid_device(struct super_block *sb,
 	if (fm->fmr_device == 0 || fm->fmr_device == UINT_MAX ||
 	    fm->fmr_device == new_encode_dev(sb->s_bdev->bd_dev))
 		return true;
-	if (EXT4_SB(sb)->s_journal_bdev &&
-	    fm->fmr_device == new_encode_dev(EXT4_SB(sb)->s_journal_bdev->bd_dev))
+	if (EXT4_SB(sb)->s_journal_bdev_handle &&
+	    fm->fmr_device ==
+	    new_encode_dev(EXT4_SB(sb)->s_journal_bdev_handle->bdev->bd_dev))
 		return true;
 	return false;
 }
@@ -647,9 +648,9 @@ int ext4_getfsmap(struct super_block *sb, struct ext4_fsmap_head *head,
 	memset(handlers, 0, sizeof(handlers));
 	handlers[0].gfd_dev = new_encode_dev(sb->s_bdev->bd_dev);
 	handlers[0].gfd_fn = ext4_getfsmap_datadev;
-	if (EXT4_SB(sb)->s_journal_bdev) {
+	if (EXT4_SB(sb)->s_journal_bdev_handle) {
 		handlers[1].gfd_dev = new_encode_dev(
-				EXT4_SB(sb)->s_journal_bdev->bd_dev);
+			EXT4_SB(sb)->s_journal_bdev_handle->bdev->bd_dev);
 		handlers[1].gfd_fn = ext4_getfsmap_logdev;
 	}
 
fs/ext4/super.c +23 -21

116 lines not on this page · raw message

Signature · 1 line
2.35.3
author reviewer
#77 lore raw

PATCH v3 22/29 ext4: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert ext4 to use bdev_open_by_dev() and pass the handle around.

---
 fs/ext4/ext4.h  |  2 +-
 fs/ext4/fsmap.c |  9 +++++----
 fs/ext4/super.c | 52 +++++++++++++++++++++++++------------------------
 3 files changed, 33 insertions(+), 30 deletions(-)
fs/ext4/ext4.h +1 -1
index 9418359b1d9d..2b3218b49bca 100644
@@ -1537,7 +1537,7 @@ struct ext4_sb_info {
 	unsigned long s_commit_interval;
 	u32 s_max_batch_time;
 	u32 s_min_batch_time;
-	struct block_device *s_journal_bdev;
+	struct bdev_handle *s_journal_bdev_handle;
 #ifdef CONFIG_QUOTA
 	/* Names of quota files with journalled quota */
 	char __rcu *s_qf_names[EXT4_MAXQUOTAS];
fs/ext4/fsmap.c +5 -4
index cdf9bfe10137..11e6f33677a2 100644
@@ -576,8 +576,9 @@ static bool ext4_getfsmap_is_valid_device(struct super_block *sb,
 	if (fm->fmr_device == 0 || fm->fmr_device == UINT_MAX ||
 	    fm->fmr_device == new_encode_dev(sb->s_bdev->bd_dev))
 		return true;
-	if (EXT4_SB(sb)->s_journal_bdev &&
-	    fm->fmr_device == new_encode_dev(EXT4_SB(sb)->s_journal_bdev->bd_dev))
+	if (EXT4_SB(sb)->s_journal_bdev_handle &&
+	    fm->fmr_device ==
+	    new_encode_dev(EXT4_SB(sb)->s_journal_bdev_handle->bdev->bd_dev))
 		return true;
 	return false;
 }
@@ -647,9 +648,9 @@ int ext4_getfsmap(struct super_block *sb, struct ext4_fsmap_head *head,
 	memset(handlers, 0, sizeof(handlers));
 	handlers[0].gfd_dev = new_encode_dev(sb->s_bdev->bd_dev);
 	handlers[0].gfd_fn = ext4_getfsmap_datadev;
-	if (EXT4_SB(sb)->s_journal_bdev) {
+	if (EXT4_SB(sb)->s_journal_bdev_handle) {
 		handlers[1].gfd_dev = new_encode_dev(
-				EXT4_SB(sb)->s_journal_bdev->bd_dev);
+			EXT4_SB(sb)->s_journal_bdev_handle->bdev->bd_dev);
 		handlers[1].gfd_fn = ext4_getfsmap_logdev;
 	}
 
fs/ext4/super.c +27 -25

137 lines not on this page · raw message

Signature · 1 line
2.35.3
author reviewer
#78 lore raw

PATCH v3 23/29 f2fs: Convert to bdev_open_by_dev/path()

In reply to Jan Kara

Convert f2fs to use bdev_open_by_dev/path() and pass the handle around.

---
 fs/f2fs/f2fs.h  |  1 +
 fs/f2fs/super.c | 17 +++++++++--------
 2 files changed, 10 insertions(+), 8 deletions(-)
fs/f2fs/f2fs.h +1 -0
index e18272ae3119..2ec6c10df636 100644
@@ -1234,6 +1234,7 @@ struct f2fs_bio_info {
 #define FDEV(i)				(sbi->devs[i])
 #define RDEV(i)				(raw_super->devs[i])
 struct f2fs_dev_info {
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 	char path[MAX_PATH_LEN];
 	unsigned int total_segments;
fs/f2fs/super.c +9 -8
index aa1f9a3a8037..885dcbd81859 100644
@@ -1561,7 +1561,7 @@ static void destroy_device_list(struct f2fs_sb_info *sbi)
 	int i;
 
 	for (i = 0; i < sbi->s_ndevs; i++) {
-		blkdev_put(FDEV(i).bdev, sbi->sb);
+		bdev_release(FDEV(i).bdev_handle);
 #ifdef CONFIG_BLK_DEV_ZONED
 		kvfree(FDEV(i).blkz_seq);
 #endif
@@ -4196,9 +4196,9 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 
 		if (max_devices == 1) {
 			/* Single zoned block device mount */
-			FDEV(0).bdev =
-				blkdev_get_by_dev(sbi->sb->s_bdev->bd_dev, mode,
-						  sbi->sb, NULL);
+			FDEV(0).bdev_handle = bdev_open_by_dev(
+					sbi->sb->s_bdev->bd_dev, mode, sbi->sb,
+					NULL);
 		} else {
 			/* Multi-device mount */
 			memcpy(FDEV(i).path, RDEV(i).path, MAX_PATH_LEN);
@@ -4216,12 +4216,13 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 					(FDEV(i).total_segments <<
 					sbi->log_blocks_per_seg) - 1;
 			}
-			FDEV(i).bdev = blkdev_get_by_path(FDEV(i).path, mode,
-							  sbi->sb, NULL);
+			FDEV(i).bdev_handle = bdev_open_by_path(FDEV(i).path,
+					mode, sbi->sb, NULL);
 		}
-		if (IS_ERR(FDEV(i).bdev))
-			return PTR_ERR(FDEV(i).bdev);
+		if (IS_ERR(FDEV(i).bdev_handle))
+			return PTR_ERR(FDEV(i).bdev_handle);
 
+		FDEV(i).bdev = FDEV(i).bdev_handle->bdev;
 		/* to release errored devices */
 		sbi->s_ndevs = i + 1;
 
Signature · 8 lines
2.35.3



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
maintainer Reviewed-by on v3 23/29
#79 lore raw

Re: [PATCH 23/29] f2fs: Convert to bdev_open_by_dev/path()

Quoted 8 lines · Jan Kara

Convert f2fs to use bdev_open_by_dev/path() and pass the handle around.

---

Looks good to me,

_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

maintainer
#80 lore raw

Re: [PATCH 23/29] f2fs: Convert to bdev_open_by_dev/path()

In reply to Jan Kara
Quoted 1 line · Jan Kara

Convert f2fs to use bdev_open_by_dev/path() and pass the handle around.

Hi Jan,

Seems it will confilct w/ below commit, could you please take a look?

https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git/commit/?h=dev&id=51bf8d3c81992ae57beeaf22df78ed7c2782af9d

Thanks,

Quoted 57 lines
---
  fs/f2fs/f2fs.h  |  1 +
  fs/f2fs/super.c | 17 +++++++++--------
  2 files changed, 10 insertions(+), 8 deletions(-)
fs/f2fs/f2fs.h +1 -0
index e18272ae3119..2ec6c10df636 100644
@@ -1234,6 +1234,7 @@ struct f2fs_bio_info {
  #define FDEV(i)				(sbi->devs[i])
  #define RDEV(i)				(raw_super->devs[i])
  struct f2fs_dev_info {
+	struct bdev_handle *bdev_handle;
  	struct block_device *bdev;
  	char path[MAX_PATH_LEN];
  	unsigned int total_segments;
fs/f2fs/super.c +6 -6
index aa1f9a3a8037..885dcbd81859 100644
@@ -1561,7 +1561,7 @@ static void destroy_device_list(struct f2fs_sb_info *sbi)
  	int i;
 
  	for (i = 0; i < sbi->s_ndevs; i++) {
-		blkdev_put(FDEV(i).bdev, sbi->sb);
+		bdev_release(FDEV(i).bdev_handle);
  #ifdef CONFIG_BLK_DEV_ZONED
  		kvfree(FDEV(i).blkz_seq);
  #endif
@@ -4196,9 +4196,9 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 
  		if (max_devices == 1) {
  			/* Single zoned block device mount */
-			FDEV(0).bdev =
-				blkdev_get_by_dev(sbi->sb->s_bdev->bd_dev, mode,
-						  sbi->sb, NULL);
+			FDEV(0).bdev_handle = bdev_open_by_dev(
+					sbi->sb->s_bdev->bd_dev, mode, sbi->sb,
+					NULL);
  		} else {
  			/* Multi-device mount */
  			memcpy(FDEV(i).path, RDEV(i).path, MAX_PATH_LEN);
@@ -4216,12 +4216,13 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
  					(FDEV(i).total_segments <<
  					sbi->log_blocks_per_seg) - 1;
  			}
-			FDEV(i).bdev = blkdev_get_by_path(FDEV(i).path, mode,
-							  sbi->sb, NULL);
+			FDEV(i).bdev_handle = bdev_open_by_path(FDEV(i).path,
+					mode, sbi->sb, NULL);

fs/f2fs/super.c static int f2fs_scan_devices(struct f2fs_sb_info *sbi) · line 4228

Quoted 10 lines
fs/f2fs/super.c +3 -2
  		}
-		if (IS_ERR(FDEV(i).bdev))
-			return PTR_ERR(FDEV(i).bdev);
+		if (IS_ERR(FDEV(i).bdev_handle))
+			return PTR_ERR(FDEV(i).bdev_handle);
 
+		FDEV(i).bdev = FDEV(i).bdev_handle->bdev;
  		/* to release errored devices */
  		sbi->s_ndevs = i + 1;
 

_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

author reviewer
#81 lore raw

Re: [PATCH 23/29] f2fs: Convert to bdev_open_by_dev/path()

Quoted 8 lines · Chao Yu
Quoted 1 line · Jan Kara

Convert f2fs to use bdev_open_by_dev/path() and pass the handle around.

Hi Jan,

Seems it will confilct w/ below commit, could you please take a look?

https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git/commit/?h=dev&id=51bf8d3c81992ae57beeaf22df78ed7c2782af9d

Yes, it will conflict. But I don't plan to rush these patches into the currently running merge window so I can just rebase after the f2fs patch gets upstream. Thanks for the heads up.

Honza

Quoted 62 lines
Quoted 62 lines
---
  fs/f2fs/f2fs.h  |  1 +
  fs/f2fs/super.c | 17 +++++++++--------
  2 files changed, 10 insertions(+), 8 deletions(-)
fs/f2fs/f2fs.h +1 -0
index e18272ae3119..2ec6c10df636 100644
@@ -1234,6 +1234,7 @@ struct f2fs_bio_info {
  #define FDEV(i)				(sbi->devs[i])
  #define RDEV(i)				(raw_super->devs[i])
  struct f2fs_dev_info {
+	struct bdev_handle *bdev_handle;
  	struct block_device *bdev;
  	char path[MAX_PATH_LEN];
  	unsigned int total_segments;
fs/f2fs/super.c +9 -8
index aa1f9a3a8037..885dcbd81859 100644
@@ -1561,7 +1561,7 @@ static void destroy_device_list(struct f2fs_sb_info *sbi)
  	int i;
  	for (i = 0; i < sbi->s_ndevs; i++) {
-		blkdev_put(FDEV(i).bdev, sbi->sb);
+		bdev_release(FDEV(i).bdev_handle);
  #ifdef CONFIG_BLK_DEV_ZONED
  		kvfree(FDEV(i).blkz_seq);
  #endif
@@ -4196,9 +4196,9 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
  		if (max_devices == 1) {
  			/* Single zoned block device mount */
-			FDEV(0).bdev =
-				blkdev_get_by_dev(sbi->sb->s_bdev->bd_dev, mode,
-						  sbi->sb, NULL);
+			FDEV(0).bdev_handle = bdev_open_by_dev(
+					sbi->sb->s_bdev->bd_dev, mode, sbi->sb,
+					NULL);
  		} else {
  			/* Multi-device mount */
  			memcpy(FDEV(i).path, RDEV(i).path, MAX_PATH_LEN);
@@ -4216,12 +4216,13 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
  					(FDEV(i).total_segments <<
  					sbi->log_blocks_per_seg) - 1;
  			}
-			FDEV(i).bdev = blkdev_get_by_path(FDEV(i).path, mode,
-							  sbi->sb, NULL);
+			FDEV(i).bdev_handle = bdev_open_by_path(FDEV(i).path,
+					mode, sbi->sb, NULL);
  		}
-		if (IS_ERR(FDEV(i).bdev))
-			return PTR_ERR(FDEV(i).bdev);
+		if (IS_ERR(FDEV(i).bdev_handle))
+			return PTR_ERR(FDEV(i).bdev_handle);
+		FDEV(i).bdev = FDEV(i).bdev_handle->bdev;
  		/* to release errored devices */
  		sbi->s_ndevs = i + 1;
Signature · 8 lines
Jan Kara <[email protected]>
SUSE Labs, CR


_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
author reviewer
#82 lore raw

PATCH v3 23/29 f2fs: Convert to bdev_open_by_dev/path()

In reply to Jan Kara

Convert f2fs to use bdev_open_by_dev/path() and pass the handle around.

---
 fs/f2fs/f2fs.h  |  1 +
 fs/f2fs/super.c | 13 +++++++------
 2 files changed, 8 insertions(+), 6 deletions(-)
fs/f2fs/f2fs.h +1 -0
index 6d688e42d89c..3878288122ee 100644
@@ -1234,6 +1234,7 @@ struct f2fs_bio_info {
 #define FDEV(i)				(sbi->devs[i])
 #define RDEV(i)				(raw_super->devs[i])
 struct f2fs_dev_info {
+	struct bdev_handle *bdev_handle;
 	struct block_device *bdev;
 	char path[MAX_PATH_LEN];
 	unsigned int total_segments;
fs/f2fs/super.c +7 -6
index a8c8232852bb..027c3ee03985 100644
@@ -1562,7 +1562,7 @@ static void destroy_device_list(struct f2fs_sb_info *sbi)
 
 	for (i = 0; i < sbi->s_ndevs; i++) {
 		if (i > 0)
-			blkdev_put(FDEV(i).bdev, sbi->sb);
+			bdev_release(FDEV(i).bdev_handle);
 #ifdef CONFIG_BLK_DEV_ZONED
 		kvfree(FDEV(i).blkz_seq);
 #endif
@@ -4198,7 +4198,7 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 
 	for (i = 0; i < max_devices; i++) {
 		if (i == 0)
-			FDEV(0).bdev = sbi->sb->s_bdev;
+			FDEV(0).bdev_handle = sbi->sb->s_bdev_handle;
 		else if (!RDEV(i).path[0])
 			break;
 
@@ -4218,13 +4218,14 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 				FDEV(i).end_blk = FDEV(i).start_blk +
 					(FDEV(i).total_segments <<
 					sbi->log_blocks_per_seg) - 1;
-				FDEV(i).bdev = blkdev_get_by_path(FDEV(i).path,
-					mode, sbi->sb, NULL);
+				FDEV(i).bdev_handle = bdev_open_by_path(
+					FDEV(i).path, mode, sbi->sb, NULL);
 			}
 		}
-		if (IS_ERR(FDEV(i).bdev))
-			return PTR_ERR(FDEV(i).bdev);
+		if (IS_ERR(FDEV(i).bdev_handle))
+			return PTR_ERR(FDEV(i).bdev_handle);
 
+		FDEV(i).bdev = FDEV(i).bdev_handle->bdev;
 		/* to release errored devices */
 		sbi->s_ndevs = i + 1;
 
Signature · 1 line
2.35.3
author reviewer
#83 lore raw

PATCH v3 24/29 jfs: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert jfs to use bdev_open_by_dev() and pass the handle around.

---
 fs/jfs/jfs_logmgr.c | 29 +++++++++++++++--------------
 fs/jfs/jfs_logmgr.h |  2 +-
 fs/jfs/jfs_mount.c  |  3 ++-
 3 files changed, 18 insertions(+), 16 deletions(-)
fs/jfs/jfs_logmgr.c +15 -14
index e855b8fde76c..c911d838b8ec 100644
@@ -1058,7 +1058,7 @@ void jfs_syncpt(struct jfs_log *log, int hard_sync)
 int lmLogOpen(struct super_block *sb)
 {
 	int rc;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct jfs_log *log;
 	struct jfs_sb_info *sbi = JFS_SBI(sb);
 
@@ -1070,7 +1070,7 @@ int lmLogOpen(struct super_block *sb)
 
 	mutex_lock(&jfs_log_mutex);
 	list_for_each_entry(log, &jfs_external_logs, journal_list) {
-		if (log->bdev->bd_dev == sbi->logdev) {
+		if (log->bdev_handle->bdev->bd_dev == sbi->logdev) {
 			if (!uuid_equal(&log->uuid, &sbi->loguuid)) {
 				jfs_warn("wrong uuid on JFS journal");
 				mutex_unlock(&jfs_log_mutex);
@@ -1100,14 +1100,14 @@ int lmLogOpen(struct super_block *sb)
 	 * file systems to log may have n-to-1 relationship;
 	 */
 
-	bdev = blkdev_get_by_dev(sbi->logdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
-				 log, NULL);
-	if (IS_ERR(bdev)) {
-		rc = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_dev(sbi->logdev,
+			BLK_OPEN_READ | BLK_OPEN_WRITE, log, NULL);
+	if (IS_ERR(bdev_handle)) {
+		rc = PTR_ERR(bdev_handle);
 		goto free;
 	}
 
-	log->bdev = bdev;
+	log->bdev_handle = bdev_handle;
 	uuid_copy(&log->uuid, &sbi->loguuid);
 
 	/*
@@ -1141,7 +1141,7 @@ int lmLogOpen(struct super_block *sb)
 	lbmLogShutdown(log);
 
       close:		/* close external log device */
-	blkdev_put(bdev, log);
+	bdev_release(bdev_handle);
 
       free:		/* free log descriptor */
 	mutex_unlock(&jfs_log_mutex);
@@ -1162,7 +1162,7 @@ static int open_inline_log(struct super_block *sb)
 	init_waitqueue_head(&log->syncwait);
 
 	set_bit(log_INLINELOG, &log->flag);
-	log->bdev = sb->s_bdev;
+	log->bdev_handle = sb->s_bdev_handle;
 	log->base = addressPXD(&JFS_SBI(sb)->logpxd);
 	log->size = lengthPXD(&JFS_SBI(sb)->logpxd) >>
 	    (L2LOGPSIZE - sb->s_blocksize_bits);
@@ -1436,7 +1436,7 @@ int lmLogClose(struct super_block *sb)
 {
 	struct jfs_sb_info *sbi = JFS_SBI(sb);
 	struct jfs_log *log = sbi->log;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	int rc = 0;
 
 	jfs_info("lmLogClose: log:0x%p", log);
@@ -1482,10 +1482,10 @@ int lmLogClose(struct super_block *sb)
 	 *	external log as separate logical volume
 	 */
 	list_del(&log->journal_list);
-	bdev = log->bdev;
+	bdev_handle = log->bdev_handle;
 	rc = lmLogShutdown(log);
 
-	blkdev_put(bdev, log);
+	bdev_release(bdev_handle);
 
 	kfree(log);
 
@@ -1972,7 +1972,7 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp)
 
 	bp->l_flag |= lbmREAD;
 
-	bio = bio_alloc(log->bdev, 1, REQ_OP_READ, GFP_NOFS);
+	bio = bio_alloc(log->bdev_handle->bdev, 1, REQ_OP_READ, GFP_NOFS);
 	bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
 	__bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
 	BUG_ON(bio->bi_iter.bi_size != LOGPSIZE);
@@ -2113,7 +2113,8 @@ static void lbmStartIO(struct lbuf * bp)
 
 	jfs_info("lbmStartIO");
 
-	bio = bio_alloc(log->bdev, 1, REQ_OP_WRITE | REQ_SYNC, GFP_NOFS);
+	bio = bio_alloc(log->bdev_handle->bdev, 1, REQ_OP_WRITE | REQ_SYNC,
+			GFP_NOFS);
 	bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
 	__bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
 	BUG_ON(bio->bi_iter.bi_size != LOGPSIZE);
fs/jfs/jfs_logmgr.h +1 -1
index 805877ce5020..84aa2d253907 100644
@@ -356,7 +356,7 @@ struct jfs_log {
 				 *    before writing syncpt.
 				 */
 	struct list_head journal_list; /* Global list */
-	struct block_device *bdev; /* 4: log lv pointer */
+	struct bdev_handle *bdev_handle; /* 4: log lv pointer */
 	int serial;		/* 4: log mount serial number */
 
 	s64 base;		/* @8: log extent address (inline log ) */
fs/jfs/jfs_mount.c +2 -1
index b83aae56a1f2..415eb65a36ff 100644
@@ -430,7 +430,8 @@ int updateSuper(struct super_block *sb, uint state)
 
 	if (state == FM_MOUNT) {
 		/* record log's dev_t and mount serial number */
-		j_sb->s_logdev = cpu_to_le32(new_encode_dev(sbi->log->bdev->bd_dev));
+		j_sb->s_logdev = cpu_to_le32(
+			new_encode_dev(sbi->log->bdev_handle->bdev->bd_dev));
 		j_sb->s_logserial = cpu_to_le32(sbi->log->serial);
 	} else if (state == FM_CLEAN) {
 		/*
Signature · 1 line
2.35.3
author reviewer
#85 lore raw

PATCH v3 24/29 jfs: Convert to bdev_open_by_dev()

In reply to Jan Kara

Convert jfs to use bdev_open_by_dev() and pass the handle around.

---
 fs/jfs/jfs_logmgr.c | 29 +++++++++++++++--------------
 fs/jfs/jfs_logmgr.h |  2 +-
 fs/jfs/jfs_mount.c  |  3 ++-
 3 files changed, 18 insertions(+), 16 deletions(-)
fs/jfs/jfs_logmgr.c +15 -14
index e855b8fde76c..c911d838b8ec 100644
@@ -1058,7 +1058,7 @@ void jfs_syncpt(struct jfs_log *log, int hard_sync)
 int lmLogOpen(struct super_block *sb)
 {
 	int rc;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	struct jfs_log *log;
 	struct jfs_sb_info *sbi = JFS_SBI(sb);
 
@@ -1070,7 +1070,7 @@ int lmLogOpen(struct super_block *sb)
 
 	mutex_lock(&jfs_log_mutex);
 	list_for_each_entry(log, &jfs_external_logs, journal_list) {
-		if (log->bdev->bd_dev == sbi->logdev) {
+		if (log->bdev_handle->bdev->bd_dev == sbi->logdev) {
 			if (!uuid_equal(&log->uuid, &sbi->loguuid)) {
 				jfs_warn("wrong uuid on JFS journal");
 				mutex_unlock(&jfs_log_mutex);
@@ -1100,14 +1100,14 @@ int lmLogOpen(struct super_block *sb)
 	 * file systems to log may have n-to-1 relationship;
 	 */
 
-	bdev = blkdev_get_by_dev(sbi->logdev, BLK_OPEN_READ | BLK_OPEN_WRITE,
-				 log, NULL);
-	if (IS_ERR(bdev)) {
-		rc = PTR_ERR(bdev);
+	bdev_handle = bdev_open_by_dev(sbi->logdev,
+			BLK_OPEN_READ | BLK_OPEN_WRITE, log, NULL);
+	if (IS_ERR(bdev_handle)) {
+		rc = PTR_ERR(bdev_handle);
 		goto free;
 	}
 
-	log->bdev = bdev;
+	log->bdev_handle = bdev_handle;
 	uuid_copy(&log->uuid, &sbi->loguuid);
 
 	/*
@@ -1141,7 +1141,7 @@ int lmLogOpen(struct super_block *sb)
 	lbmLogShutdown(log);
 
       close:		/* close external log device */
-	blkdev_put(bdev, log);
+	bdev_release(bdev_handle);
 
       free:		/* free log descriptor */
 	mutex_unlock(&jfs_log_mutex);
@@ -1162,7 +1162,7 @@ static int open_inline_log(struct super_block *sb)
 	init_waitqueue_head(&log->syncwait);
 
 	set_bit(log_INLINELOG, &log->flag);
-	log->bdev = sb->s_bdev;
+	log->bdev_handle = sb->s_bdev_handle;
 	log->base = addressPXD(&JFS_SBI(sb)->logpxd);
 	log->size = lengthPXD(&JFS_SBI(sb)->logpxd) >>
 	    (L2LOGPSIZE - sb->s_blocksize_bits);
@@ -1436,7 +1436,7 @@ int lmLogClose(struct super_block *sb)
 {
 	struct jfs_sb_info *sbi = JFS_SBI(sb);
 	struct jfs_log *log = sbi->log;
-	struct block_device *bdev;
+	struct bdev_handle *bdev_handle;
 	int rc = 0;
 
 	jfs_info("lmLogClose: log:0x%p", log);
@@ -1482,10 +1482,10 @@ int lmLogClose(struct super_block *sb)
 	 *	external log as separate logical volume
 	 */
 	list_del(&log->journal_list);
-	bdev = log->bdev;
+	bdev_handle = log->bdev_handle;
 	rc = lmLogShutdown(log);
 
-	blkdev_put(bdev, log);
+	bdev_release(bdev_handle);
 
 	kfree(log);
 
@@ -1972,7 +1972,7 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp)
 
 	bp->l_flag |= lbmREAD;
 
-	bio = bio_alloc(log->bdev, 1, REQ_OP_READ, GFP_NOFS);
+	bio = bio_alloc(log->bdev_handle->bdev, 1, REQ_OP_READ, GFP_NOFS);
 	bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
 	__bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
 	BUG_ON(bio->bi_iter.bi_size != LOGPSIZE);
@@ -2113,7 +2113,8 @@ static void lbmStartIO(struct lbuf * bp)
 
 	jfs_info("lbmStartIO");
 
-	bio = bio_alloc(log->bdev, 1, REQ_OP_WRITE | REQ_SYNC, GFP_NOFS);
+	bio = bio_alloc(log->bdev_handle->bdev, 1, REQ_OP_WRITE | REQ_SYNC,
+			GFP_NOFS);
 	bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
 	__bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
 	BUG_ON(bio->bi_iter.bi_size != LOGPSIZE);
fs/jfs/jfs_logmgr.h +1 -1
index 805877ce5020..84aa2d253907 100644
@@ -356,7 +356,7 @@ struct jfs_log {
 				 *    before writing syncpt.
 				 */
 	struct list_head journal_list; /* Global list */
-	struct block_device *bdev; /* 4: log lv pointer */
+	struct bdev_handle *bdev_handle; /* 4: log lv pointer */
 	int serial;		/* 4: log mount serial number */
 
 	s64 base;		/* @8: log extent address (inline log ) */
fs/jfs/jfs_mount.c +2 -1
index b83aae56a1f2..415eb65a36ff 100644
@@ -430,7 +430,8 @@ int updateSuper(struct super_block *sb, uint state)
 
 	if (state == FM_MOUNT) {
 		/* record log's dev_t and mount serial number */
-		j_sb->s_logdev = cpu_to_le32(new_encode_dev(sbi->log->bdev->bd_dev));
+		j_sb->s_logdev = cpu_to_le32(
+			new_encode_dev(sbi->log->bdev_handle->bdev->bd_dev));
 		j_sb->s_logserial = cpu_to_le32(sbi->log->serial);
 	} else if (state == FM_CLEAN) {
 		/*
Signature · 1 line
2.35.3
author reviewer
#86 lore raw

PATCH v3 25/29 nfs/blocklayout: Convert to use bdev_open_by_dev/path()

In reply to Jan Kara

Convert block device handling to use bdev_open_by_dev/path() and pass the handle around.

---
 fs/nfs/blocklayout/blocklayout.h |  2 +-
 fs/nfs/blocklayout/dev.c         | 76 ++++++++++++++++----------------
 2 files changed, 38 insertions(+), 40 deletions(-)
fs/nfs/blocklayout/blocklayout.h +1 -1
index 716bc75e9ed2..b4294a8aa2d4 100644
@@ -108,7 +108,7 @@ struct pnfs_block_dev {
 	struct pnfs_block_dev		*children;
 	u64				chunk_size;
 
-	struct block_device		*bdev;
+	struct bdev_handle		*bdev_handle;
 	u64				disk_offset;
 
 	u64				pr_key;
fs/nfs/blocklayout/dev.c +37 -39

160 lines not on this page · raw message

Signature · 1 line
2.35.3
maintainer Reviewed-by on v3 25/29
#87 lore raw

Re: [PATCH 25/29] nfs/blocklayout: Convert to use bdev_open_by_dev/path()

Quoted 9 lines · Jan Kara

Convert block device handling to use bdev_open_by_dev/path() and pass the handle around.

---

Looks good to me,

author reviewer
#88 lore raw

PATCH v3 25/29 nfs/blocklayout: Convert to use bdev_open_by_dev/path()

In reply to Jan Kara

Convert block device handling to use bdev_open_by_dev/path() and pass the handle around.

---
 fs/nfs/blocklayout/blocklayout.h |  2 +-
 fs/nfs/blocklayout/dev.c         | 76 ++++++++++++++++----------------
 2 files changed, 38 insertions(+), 40 deletions(-)
fs/nfs/blocklayout/blocklayout.h +1 -1
index 716bc75e9ed2..b4294a8aa2d4 100644
@@ -108,7 +108,7 @@ struct pnfs_block_dev {
 	struct pnfs_block_dev		*children;
 	u64				chunk_size;
 
-	struct block_device		*bdev;
+	struct bdev_handle		*bdev_handle;
 	u64				disk_offset;
 
 	u64				pr_key;
fs/nfs/blocklayout/dev.c +37 -39

160 lines not on this page · raw message

Signature · 1 line
2.35.3
author reviewer
#89 lore raw

PATCH v3 26/29 ocfs2: Convert to use bdev_open_by_dev()

In reply to Jan Kara

Convert ocfs2 heartbeat code to use bdev_open_by_dev() and pass the handle around.

---
 fs/ocfs2/cluster/heartbeat.c | 81 ++++++++++++++++++++----------------
 1 file changed, 45 insertions(+), 36 deletions(-)
fs/ocfs2/cluster/heartbeat.c +45 -36
index 21472e3ed182..4d7efefa98c5 100644
@@ -213,7 +213,7 @@ struct o2hb_region {
 	unsigned int		hr_num_pages;
 
 	struct page             **hr_slot_data;
-	struct block_device	*hr_bdev;
+	struct bdev_handle	*hr_bdev_handle;
 	struct o2hb_disk_slot	*hr_slots;
 
 	/* live node map of this region */
@@ -261,6 +261,11 @@ struct o2hb_region {
 	int			hr_last_hb_status;
 };
 
+static inline struct block_device *reg_bdev(struct o2hb_region *reg)
+{
+	return reg->hr_bdev_handle ? reg->hr_bdev_handle->bdev : NULL;
+}
+
 struct o2hb_bio_wait_ctxt {
 	atomic_t          wc_num_reqs;
 	struct completion wc_io_complete;
@@ -286,7 +291,7 @@ static void o2hb_write_timeout(struct work_struct *work)
 			     hr_write_timeout_work.work);
 
 	mlog(ML_ERROR, "Heartbeat write timeout to device %pg after %u "
-	     "milliseconds\n", reg->hr_bdev,
+	     "milliseconds\n", reg_bdev(reg),
 	     jiffies_to_msecs(jiffies - reg->hr_last_timeout_start));
 
 	if (o2hb_global_heartbeat_active()) {
@@ -383,7 +388,7 @@ static void o2hb_nego_timeout(struct work_struct *work)
 		if (!test_bit(master_node, reg->hr_nego_node_bitmap)) {
 			printk(KERN_NOTICE "o2hb: node %d hb write hung for %ds on region %s (%pg).\n",
 				o2nm_this_node(), O2HB_NEGO_TIMEOUT_MS/1000,
-				config_item_name(&reg->hr_item), reg->hr_bdev);
+				config_item_name(&reg->hr_item), reg_bdev(reg));
 			set_bit(master_node, reg->hr_nego_node_bitmap);
 		}
 		if (!bitmap_equal(reg->hr_nego_node_bitmap, live_node_bitmap,
@@ -398,7 +403,8 @@ static void o2hb_nego_timeout(struct work_struct *work)
 		}
 
 		printk(KERN_NOTICE "o2hb: all nodes hb write hung, maybe region %s (%pg) is down.\n",
-			config_item_name(&reg->hr_item), reg->hr_bdev);
+			config_item_name(&reg->hr_item),
+			reg_bdev(reg));
 		/* approve negotiate timeout request. */
 		o2hb_arm_timeout(reg);
 
@@ -419,7 +425,7 @@ static void o2hb_nego_timeout(struct work_struct *work)
 		/* negotiate timeout with master node. */
 		printk(KERN_NOTICE "o2hb: node %d hb write hung for %ds on region %s (%pg), negotiate timeout with node %d.\n",
 			o2nm_this_node(), O2HB_NEGO_TIMEOUT_MS/1000, config_item_name(&reg->hr_item),
-			reg->hr_bdev, master_node);
+			reg_bdev(reg), master_node);
 		ret = o2hb_send_nego_msg(reg->hr_key, O2HB_NEGO_TIMEOUT_MSG,
 				master_node);
 		if (ret)
@@ -436,7 +442,8 @@ static int o2hb_nego_timeout_handler(struct o2net_msg *msg, u32 len, void *data,
 
 	nego_msg = (struct o2hb_nego_msg *)msg->buf;
 	printk(KERN_NOTICE "o2hb: receive negotiate timeout message from node %d on region %s (%pg).\n",
-		nego_msg->node_num, config_item_name(&reg->hr_item), reg->hr_bdev);
+		nego_msg->node_num, config_item_name(&reg->hr_item),
+		reg_bdev(reg));
 	if (nego_msg->node_num < O2NM_MAX_NODES)
 		set_bit(nego_msg->node_num, reg->hr_nego_node_bitmap);
 	else
@@ -451,7 +458,7 @@ static int o2hb_nego_approve_handler(struct o2net_msg *msg, u32 len, void *data,
 	struct o2hb_region *reg = data;
 
 	printk(KERN_NOTICE "o2hb: negotiate timeout approved by master node on region %s (%pg).\n",
-		config_item_name(&reg->hr_item), reg->hr_bdev);
+		config_item_name(&reg->hr_item), reg_bdev(reg));
 	o2hb_arm_timeout(reg);
 	return 0;
 }
@@ -515,7 +522,7 @@ static struct bio *o2hb_setup_one_bio(struct o2hb_region *reg,
 	 * GFP_KERNEL that the local node can get fenced. It would be
 	 * nicest if we could pre-allocate these bios and avoid this
 	 * all together. */
-	bio = bio_alloc(reg->hr_bdev, 16, opf, GFP_ATOMIC);
+	bio = bio_alloc(reg_bdev(reg), 16, opf, GFP_ATOMIC);
 	if (!bio) {
 		mlog(ML_ERROR, "Could not alloc slots BIO!\n");
 		bio = ERR_PTR(-ENOMEM);
@@ -687,7 +694,7 @@ static int o2hb_check_own_slot(struct o2hb_region *reg)
 		errstr = ERRSTR3;
 
 	mlog(ML_ERROR, "%s (%pg): expected(%u:0x%llx, 0x%llx), "
-	     "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_bdev,
+	     "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg_bdev(reg),
 	     slot->ds_node_num, (unsigned long long)slot->ds_last_generation,
 	     (unsigned long long)slot->ds_last_time, hb_block->hb_node,
 	     (unsigned long long)le64_to_cpu(hb_block->hb_generation),
@@ -861,7 +868,7 @@ static void o2hb_set_quorum_device(struct o2hb_region *reg)
 		goto unlock;
 
 	printk(KERN_NOTICE "o2hb: Region %s (%pg) is now a quorum device\n",
-	       config_item_name(&reg->hr_item), reg->hr_bdev);
+	       config_item_name(&reg->hr_item), reg_bdev(reg));
 
 	set_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
 
@@ -920,7 +927,7 @@ static int o2hb_check_slot(struct o2hb_region *reg,
 		 * consider it a transient miss but don't populate any
 		 * other values as they may be junk. */
 		mlog(ML_ERROR, "Node %d has written a bad crc to %pg\n",
-		     slot->ds_node_num, reg->hr_bdev);
+		     slot->ds_node_num, reg_bdev(reg));
 		o2hb_dump_slot(hb_block);
 
 		slot->ds_equal_samples++;
@@ -1003,8 +1010,8 @@ static int o2hb_check_slot(struct o2hb_region *reg,
 			     "of %u ms, but our count is %u ms.\n"
 			     "Please double check your configuration values "
 			     "for 'O2CB_HEARTBEAT_THRESHOLD'\n",
-			     slot->ds_node_num, reg->hr_bdev, slot_dead_ms,
-			     dead_ms);
+			     slot->ds_node_num, reg_bdev(reg),
+			     slot_dead_ms, dead_ms);
 		}
 		goto out;
 	}
@@ -1143,7 +1150,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 		 * can't be sure that the new block ever made it to
 		 * disk */
 		mlog(ML_ERROR, "Write error %d on device \"%pg\"\n",
-		     write_wc.wc_error, reg->hr_bdev);
+		     write_wc.wc_error, reg_bdev(reg));
 		ret = write_wc.wc_error;
 		goto bail;
 	}
@@ -1169,7 +1176,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 			printk(KERN_NOTICE "o2hb: Unable to stabilize "
 			       "heartbeat on region %s (%pg)\n",
 			       config_item_name(&reg->hr_item),
-			       reg->hr_bdev);
+			       reg_bdev(reg));
 			atomic_set(&reg->hr_steady_iterations, 0);
 			reg->hr_aborted_start = 1;
 			wake_up(&o2hb_steady_queue);
@@ -1489,7 +1496,7 @@ static void o2hb_region_release(struct config_item *item)
 	struct page *page;
 	struct o2hb_region *reg = to_o2hb_region(item);
 
-	mlog(ML_HEARTBEAT, "hb region release (%pg)\n", reg->hr_bdev);
+	mlog(ML_HEARTBEAT, "hb region release (%pg)\n", reg_bdev(reg));
 
 	kfree(reg->hr_tmp_block);
 
@@ -1502,8 +1509,8 @@ static void o2hb_region_release(struct config_item *item)
 		kfree(reg->hr_slot_data);
 	}
 
-	if (reg->hr_bdev)
-		blkdev_put(reg->hr_bdev, NULL);
+	if (reg->hr_bdev_handle)
+		bdev_release(reg->hr_bdev_handle);
 
 	kfree(reg->hr_slots);
 
@@ -1562,7 +1569,7 @@ static ssize_t o2hb_region_block_bytes_store(struct config_item *item,
 	unsigned long block_bytes;
 	unsigned int block_bits;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		return -EINVAL;
 
 	status = o2hb_read_block_input(reg, page, &block_bytes,
@@ -1591,7 +1598,7 @@ static ssize_t o2hb_region_start_block_store(struct config_item *item,
 	char *p = (char *)page;
 	ssize_t ret;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		return -EINVAL;
 
 	ret = kstrtoull(p, 0, &tmp);
@@ -1616,7 +1623,7 @@ static ssize_t o2hb_region_blocks_store(struct config_item *item,
 	unsigned long tmp;
 	char *p = (char *)page;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		return -EINVAL;
 
 	tmp = simple_strtoul(p, &p, 0);
@@ -1635,8 +1642,8 @@ static ssize_t o2hb_region_dev_show(struct config_item *item, char *page)
 {
 	unsigned int ret = 0;
 
-	if (to_o2hb_region(item)->hr_bdev)
-		ret = sprintf(page, "%pg\n", to_o2hb_region(item)->hr_bdev);
+	if (to_o2hb_region(item)->hr_bdev_handle)
+		ret = sprintf(page, "%pg\n", reg_bdev(to_o2hb_region(item)));
 
 	return ret;
 }
@@ -1745,7 +1752,10 @@ static int o2hb_populate_slot_data(struct o2hb_region *reg)
 	return ret;
 }
 
-/* this is acting as commit; we set up all of hr_bdev and hr_task or nothing */
+/*
+ * this is acting as commit; we set up all of hr_bdev_handle and hr_task or
+ * nothing
+ */
 static ssize_t o2hb_region_dev_store(struct config_item *item,
 				     const char *page,
 				     size_t count)
@@ -1759,7 +1769,7 @@ static ssize_t o2hb_region_dev_store(struct config_item *item,
 	ssize_t ret = -EINVAL;
 	int live_threshold;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		goto out;
 
 	/* We can't heartbeat without having had our node number
@@ -1785,16 +1795,15 @@ static ssize_t o2hb_region_dev_store(struct config_item *item,
 	if (!S_ISBLK(f.file->f_mapping->host->i_mode))
 		goto out2;
 
-	reg->hr_bdev = blkdev_get_by_dev(f.file->f_mapping->host->i_rdev,
-					 BLK_OPEN_WRITE | BLK_OPEN_READ, NULL,
-					 NULL);
-	if (IS_ERR(reg->hr_bdev)) {
-		ret = PTR_ERR(reg->hr_bdev);
-		reg->hr_bdev = NULL;
+	reg->hr_bdev_handle = bdev_open_by_dev(f.file->f_mapping->host->i_rdev,
+			BLK_OPEN_WRITE | BLK_OPEN_READ, NULL, NULL);
+	if (IS_ERR(reg->hr_bdev_handle)) {
+		ret = PTR_ERR(reg->hr_bdev_handle);
+		reg->hr_bdev_handle = NULL;
 		goto out2;
 	}
 
-	sectsize = bdev_logical_block_size(reg->hr_bdev);
+	sectsize = bdev_logical_block_size(reg_bdev(reg));
 	if (sectsize != reg->hr_block_bytes) {
 		mlog(ML_ERROR,
 		     "blocksize %u incorrect for device, expected %d",
@@ -1890,12 +1899,12 @@ static ssize_t o2hb_region_dev_store(struct config_item *item,
 
 	if (hb_task && o2hb_global_heartbeat_active())
 		printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%pg)\n",
-		       config_item_name(&reg->hr_item), reg->hr_bdev);
+		       config_item_name(&reg->hr_item), reg_bdev(reg));
 
 out3:
 	if (ret < 0) {
-		blkdev_put(reg->hr_bdev, NULL);
-		reg->hr_bdev = NULL;
+		bdev_release(reg->hr_bdev_handle);
+		reg->hr_bdev_handle = NULL;
 	}
 out2:
 	fdput(f);
@@ -2085,7 +2094,7 @@ static void o2hb_heartbeat_group_drop_item(struct config_group *group,
 		printk(KERN_NOTICE "o2hb: Heartbeat %s on region %s (%pg)\n",
 		       ((atomic_read(&reg->hr_steady_iterations) == 0) ?
 			"stopped" : "start aborted"), config_item_name(item),
-		       reg->hr_bdev);
+		       reg_bdev(reg));
 	}
 
 	/*
Signature · 1 line
2.35.3
maintainer Reviewed-by on v3 26/29
#90 lore raw

Re: [PATCH 26/29] ocfs2: Convert to use bdev_open_by_dev()

Quoted 9 lines · Jan Kara

Convert ocfs2 heartbeat code to use bdev_open_by_dev() and pass the handle around.

---

Looks good to me,

author reviewer
#91 lore raw

PATCH v3 26/29 ocfs2: Convert to use bdev_open_by_dev()

In reply to Jan Kara

Convert ocfs2 heartbeat code to use bdev_open_by_dev() and pass the handle around.

---
 fs/ocfs2/cluster/heartbeat.c | 81 ++++++++++++++++++++----------------
 1 file changed, 45 insertions(+), 36 deletions(-)
fs/ocfs2/cluster/heartbeat.c +45 -36
index 21472e3ed182..4d7efefa98c5 100644
@@ -213,7 +213,7 @@ struct o2hb_region {
 	unsigned int		hr_num_pages;
 
 	struct page             **hr_slot_data;
-	struct block_device	*hr_bdev;
+	struct bdev_handle	*hr_bdev_handle;
 	struct o2hb_disk_slot	*hr_slots;
 
 	/* live node map of this region */
@@ -261,6 +261,11 @@ struct o2hb_region {
 	int			hr_last_hb_status;
 };
 
+static inline struct block_device *reg_bdev(struct o2hb_region *reg)
+{
+	return reg->hr_bdev_handle ? reg->hr_bdev_handle->bdev : NULL;
+}
+
 struct o2hb_bio_wait_ctxt {
 	atomic_t          wc_num_reqs;
 	struct completion wc_io_complete;
@@ -286,7 +291,7 @@ static void o2hb_write_timeout(struct work_struct *work)
 			     hr_write_timeout_work.work);
 
 	mlog(ML_ERROR, "Heartbeat write timeout to device %pg after %u "
-	     "milliseconds\n", reg->hr_bdev,
+	     "milliseconds\n", reg_bdev(reg),
 	     jiffies_to_msecs(jiffies - reg->hr_last_timeout_start));
 
 	if (o2hb_global_heartbeat_active()) {
@@ -383,7 +388,7 @@ static void o2hb_nego_timeout(struct work_struct *work)
 		if (!test_bit(master_node, reg->hr_nego_node_bitmap)) {
 			printk(KERN_NOTICE "o2hb: node %d hb write hung for %ds on region %s (%pg).\n",
 				o2nm_this_node(), O2HB_NEGO_TIMEOUT_MS/1000,
-				config_item_name(&reg->hr_item), reg->hr_bdev);
+				config_item_name(&reg->hr_item), reg_bdev(reg));
 			set_bit(master_node, reg->hr_nego_node_bitmap);
 		}
 		if (!bitmap_equal(reg->hr_nego_node_bitmap, live_node_bitmap,
@@ -398,7 +403,8 @@ static void o2hb_nego_timeout(struct work_struct *work)
 		}
 
 		printk(KERN_NOTICE "o2hb: all nodes hb write hung, maybe region %s (%pg) is down.\n",
-			config_item_name(&reg->hr_item), reg->hr_bdev);
+			config_item_name(&reg->hr_item),
+			reg_bdev(reg));
 		/* approve negotiate timeout request. */
 		o2hb_arm_timeout(reg);
 
@@ -419,7 +425,7 @@ static void o2hb_nego_timeout(struct work_struct *work)
 		/* negotiate timeout with master node. */
 		printk(KERN_NOTICE "o2hb: node %d hb write hung for %ds on region %s (%pg), negotiate timeout with node %d.\n",
 			o2nm_this_node(), O2HB_NEGO_TIMEOUT_MS/1000, config_item_name(&reg->hr_item),
-			reg->hr_bdev, master_node);
+			reg_bdev(reg), master_node);
 		ret = o2hb_send_nego_msg(reg->hr_key, O2HB_NEGO_TIMEOUT_MSG,
 				master_node);
 		if (ret)
@@ -436,7 +442,8 @@ static int o2hb_nego_timeout_handler(struct o2net_msg *msg, u32 len, void *data,
 
 	nego_msg = (struct o2hb_nego_msg *)msg->buf;
 	printk(KERN_NOTICE "o2hb: receive negotiate timeout message from node %d on region %s (%pg).\n",
-		nego_msg->node_num, config_item_name(&reg->hr_item), reg->hr_bdev);
+		nego_msg->node_num, config_item_name(&reg->hr_item),
+		reg_bdev(reg));
 	if (nego_msg->node_num < O2NM_MAX_NODES)
 		set_bit(nego_msg->node_num, reg->hr_nego_node_bitmap);
 	else
@@ -451,7 +458,7 @@ static int o2hb_nego_approve_handler(struct o2net_msg *msg, u32 len, void *data,
 	struct o2hb_region *reg = data;
 
 	printk(KERN_NOTICE "o2hb: negotiate timeout approved by master node on region %s (%pg).\n",
-		config_item_name(&reg->hr_item), reg->hr_bdev);
+		config_item_name(&reg->hr_item), reg_bdev(reg));
 	o2hb_arm_timeout(reg);
 	return 0;
 }
@@ -515,7 +522,7 @@ static struct bio *o2hb_setup_one_bio(struct o2hb_region *reg,
 	 * GFP_KERNEL that the local node can get fenced. It would be
 	 * nicest if we could pre-allocate these bios and avoid this
 	 * all together. */
-	bio = bio_alloc(reg->hr_bdev, 16, opf, GFP_ATOMIC);
+	bio = bio_alloc(reg_bdev(reg), 16, opf, GFP_ATOMIC);
 	if (!bio) {
 		mlog(ML_ERROR, "Could not alloc slots BIO!\n");
 		bio = ERR_PTR(-ENOMEM);
@@ -687,7 +694,7 @@ static int o2hb_check_own_slot(struct o2hb_region *reg)
 		errstr = ERRSTR3;
 
 	mlog(ML_ERROR, "%s (%pg): expected(%u:0x%llx, 0x%llx), "
-	     "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_bdev,
+	     "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg_bdev(reg),
 	     slot->ds_node_num, (unsigned long long)slot->ds_last_generation,
 	     (unsigned long long)slot->ds_last_time, hb_block->hb_node,
 	     (unsigned long long)le64_to_cpu(hb_block->hb_generation),
@@ -861,7 +868,7 @@ static void o2hb_set_quorum_device(struct o2hb_region *reg)
 		goto unlock;
 
 	printk(KERN_NOTICE "o2hb: Region %s (%pg) is now a quorum device\n",
-	       config_item_name(&reg->hr_item), reg->hr_bdev);
+	       config_item_name(&reg->hr_item), reg_bdev(reg));
 
 	set_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
 
@@ -920,7 +927,7 @@ static int o2hb_check_slot(struct o2hb_region *reg,
 		 * consider it a transient miss but don't populate any
 		 * other values as they may be junk. */
 		mlog(ML_ERROR, "Node %d has written a bad crc to %pg\n",
-		     slot->ds_node_num, reg->hr_bdev);
+		     slot->ds_node_num, reg_bdev(reg));
 		o2hb_dump_slot(hb_block);
 
 		slot->ds_equal_samples++;
@@ -1003,8 +1010,8 @@ static int o2hb_check_slot(struct o2hb_region *reg,
 			     "of %u ms, but our count is %u ms.\n"
 			     "Please double check your configuration values "
 			     "for 'O2CB_HEARTBEAT_THRESHOLD'\n",
-			     slot->ds_node_num, reg->hr_bdev, slot_dead_ms,
-			     dead_ms);
+			     slot->ds_node_num, reg_bdev(reg),
+			     slot_dead_ms, dead_ms);
 		}
 		goto out;
 	}
@@ -1143,7 +1150,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 		 * can't be sure that the new block ever made it to
 		 * disk */
 		mlog(ML_ERROR, "Write error %d on device \"%pg\"\n",
-		     write_wc.wc_error, reg->hr_bdev);
+		     write_wc.wc_error, reg_bdev(reg));
 		ret = write_wc.wc_error;
 		goto bail;
 	}
@@ -1169,7 +1176,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 			printk(KERN_NOTICE "o2hb: Unable to stabilize "
 			       "heartbeat on region %s (%pg)\n",
 			       config_item_name(&reg->hr_item),
-			       reg->hr_bdev);
+			       reg_bdev(reg));
 			atomic_set(&reg->hr_steady_iterations, 0);
 			reg->hr_aborted_start = 1;
 			wake_up(&o2hb_steady_queue);
@@ -1489,7 +1496,7 @@ static void o2hb_region_release(struct config_item *item)
 	struct page *page;
 	struct o2hb_region *reg = to_o2hb_region(item);
 
-	mlog(ML_HEARTBEAT, "hb region release (%pg)\n", reg->hr_bdev);
+	mlog(ML_HEARTBEAT, "hb region release (%pg)\n", reg_bdev(reg));
 
 	kfree(reg->hr_tmp_block);
 
@@ -1502,8 +1509,8 @@ static void o2hb_region_release(struct config_item *item)
 		kfree(reg->hr_slot_data);
 	}
 
-	if (reg->hr_bdev)
-		blkdev_put(reg->hr_bdev, NULL);
+	if (reg->hr_bdev_handle)
+		bdev_release(reg->hr_bdev_handle);
 
 	kfree(reg->hr_slots);
 
@@ -1562,7 +1569,7 @@ static ssize_t o2hb_region_block_bytes_store(struct config_item *item,
 	unsigned long block_bytes;
 	unsigned int block_bits;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		return -EINVAL;
 
 	status = o2hb_read_block_input(reg, page, &block_bytes,
@@ -1591,7 +1598,7 @@ static ssize_t o2hb_region_start_block_store(struct config_item *item,
 	char *p = (char *)page;
 	ssize_t ret;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		return -EINVAL;
 
 	ret = kstrtoull(p, 0, &tmp);
@@ -1616,7 +1623,7 @@ static ssize_t o2hb_region_blocks_store(struct config_item *item,
 	unsigned long tmp;
 	char *p = (char *)page;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		return -EINVAL;
 
 	tmp = simple_strtoul(p, &p, 0);
@@ -1635,8 +1642,8 @@ static ssize_t o2hb_region_dev_show(struct config_item *item, char *page)
 {
 	unsigned int ret = 0;
 
-	if (to_o2hb_region(item)->hr_bdev)
-		ret = sprintf(page, "%pg\n", to_o2hb_region(item)->hr_bdev);
+	if (to_o2hb_region(item)->hr_bdev_handle)
+		ret = sprintf(page, "%pg\n", reg_bdev(to_o2hb_region(item)));
 
 	return ret;
 }
@@ -1745,7 +1752,10 @@ static int o2hb_populate_slot_data(struct o2hb_region *reg)
 	return ret;
 }
 
-/* this is acting as commit; we set up all of hr_bdev and hr_task or nothing */
+/*
+ * this is acting as commit; we set up all of hr_bdev_handle and hr_task or
+ * nothing
+ */
 static ssize_t o2hb_region_dev_store(struct config_item *item,
 				     const char *page,
 				     size_t count)
@@ -1759,7 +1769,7 @@ static ssize_t o2hb_region_dev_store(struct config_item *item,
 	ssize_t ret = -EINVAL;
 	int live_threshold;
 
-	if (reg->hr_bdev)
+	if (reg->hr_bdev_handle)
 		goto out;
 
 	/* We can't heartbeat without having had our node number
@@ -1785,16 +1795,15 @@ static ssize_t o2hb_region_dev_store(struct config_item *item,
 	if (!S_ISBLK(f.file->f_mapping->host->i_mode))
 		goto out2;
 
-	reg->hr_bdev = blkdev_get_by_dev(f.file->f_mapping->host->i_rdev,
-					 BLK_OPEN_WRITE | BLK_OPEN_READ, NULL,
-					 NULL);
-	if (IS_ERR(reg->hr_bdev)) {
-		ret = PTR_ERR(reg->hr_bdev);
-		reg->hr_bdev = NULL;
+	reg->hr_bdev_handle = bdev_open_by_dev(f.file->f_mapping->host->i_rdev,
+			BLK_OPEN_WRITE | BLK_OPEN_READ, NULL, NULL);
+	if (IS_ERR(reg->hr_bdev_handle)) {
+		ret = PTR_ERR(reg->hr_bdev_handle);
+		reg->hr_bdev_handle = NULL;
 		goto out2;
 	}
 
-	sectsize = bdev_logical_block_size(reg->hr_bdev);
+	sectsize = bdev_logical_block_size(reg_bdev(reg));
 	if (sectsize != reg->hr_block_bytes) {
 		mlog(ML_ERROR,
 		     "blocksize %u incorrect for device, expected %d",
@@ -1890,12 +1899,12 @@ static ssize_t o2hb_region_dev_store(struct config_item *item,
 
 	if (hb_task && o2hb_global_heartbeat_active())
 		printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%pg)\n",
-		       config_item_name(&reg->hr_item), reg->hr_bdev);
+		       config_item_name(&reg->hr_item), reg_bdev(reg));
 
 out3:
 	if (ret < 0) {
-		blkdev_put(reg->hr_bdev, NULL);
-		reg->hr_bdev = NULL;
+		bdev_release(reg->hr_bdev_handle);
+		reg->hr_bdev_handle = NULL;
 	}
 out2:
 	fdput(f);
@@ -2085,7 +2094,7 @@ static void o2hb_heartbeat_group_drop_item(struct config_group *group,
 		printk(KERN_NOTICE "o2hb: Heartbeat %s on region %s (%pg)\n",
 		       ((atomic_read(&reg->hr_steady_iterations) == 0) ?
 			"stopped" : "start aborted"), config_item_name(item),
-		       reg->hr_bdev);
+		       reg_bdev(reg));
 	}
 
 	/*
Signature · 1 line
2.35.3
author reviewer
#92 lore raw

PATCH v3 27/29 reiserfs: Convert to bdev_open_by_dev/path()

In reply to Jan Kara

Convert reiserfs to use bdev_open_by_dev/path() and pass the handle around.

---
 fs/reiserfs/journal.c  | 56 +++++++++++++++++++-----------------------
 fs/reiserfs/procfs.c   |  2 +-
 fs/reiserfs/reiserfs.h | 11 ++++++---
 3 files changed, 33 insertions(+), 36 deletions(-)
fs/reiserfs/journal.c +25 -31
index 015bfe4e4524..171c912af50f 100644
@@ -90,8 +90,7 @@ static int flush_commit_list(struct super_block *s,
 static int can_dirty(struct reiserfs_journal_cnode *cn);
 static int journal_join(struct reiserfs_transaction_handle *th,
 			struct super_block *sb);
-static void release_journal_dev(struct super_block *super,
-			       struct reiserfs_journal *journal);
+static void release_journal_dev(struct reiserfs_journal *journal);
 static void dirty_one_transaction(struct super_block *s,
 				 struct reiserfs_journal_list *jl);
 static void flush_async_commits(struct work_struct *work);
@@ -1893,7 +1892,7 @@ static void free_journal_ram(struct super_block *sb)
 	 * j_header_bh is on the journal dev, make sure
 	 * not to release the journal dev until we brelse j_header_bh
 	 */
-	release_journal_dev(sb, journal);
+	release_journal_dev(journal);
 	vfree(journal);
 }
 
@@ -2387,7 +2386,7 @@ static int journal_read(struct super_block *sb)
 
 	cur_dblock = SB_ONDISK_JOURNAL_1st_BLOCK(sb);
 	reiserfs_info(sb, "checking transaction log (%pg)\n",
-		      journal->j_dev_bd);
+		      journal->j_bdev_handle->bdev);
 	start = ktime_get_seconds();
 
 	/*
@@ -2448,7 +2447,7 @@ static int journal_read(struct super_block *sb)
 		 * device and journal device to be the same
 		 */
 		d_bh =
-		    reiserfs_breada(journal->j_dev_bd, cur_dblock,
+		    reiserfs_breada(journal->j_bdev_handle->bdev, cur_dblock,
 				    sb->s_blocksize,
 				    SB_ONDISK_JOURNAL_1st_BLOCK(sb) +
 				    SB_ONDISK_JOURNAL_SIZE(sb));
@@ -2587,17 +2586,11 @@ static void journal_list_init(struct super_block *sb)
 	SB_JOURNAL(sb)->j_current_jl = alloc_journal_list(sb);
 }
 
-static void release_journal_dev(struct super_block *super,
-			       struct reiserfs_journal *journal)
+static void release_journal_dev(struct reiserfs_journal *journal)
 {
-	if (journal->j_dev_bd != NULL) {
-		void *holder = NULL;
-
-		if (journal->j_dev_bd->bd_dev != super->s_dev)
-			holder = journal;
-
-		blkdev_put(journal->j_dev_bd, holder);
-		journal->j_dev_bd = NULL;
+	if (journal->j_bdev_handle) {
+		bdev_release(journal->j_bdev_handle);
+		journal->j_bdev_handle = NULL;
 	}
 }
 
@@ -2612,7 +2605,7 @@ static int journal_init_dev(struct super_block *super,
 
 	result = 0;
 
-	journal->j_dev_bd = NULL;
+	journal->j_bdev_handle = NULL;
 	jdev = SB_ONDISK_JOURNAL_DEVICE(super) ?
 	    new_decode_dev(SB_ONDISK_JOURNAL_DEVICE(super)) : super->s_dev;
 
@@ -2623,36 +2616,37 @@ static int journal_init_dev(struct super_block *super,
 	if ((!jdev_name || !jdev_name[0])) {
 		if (jdev == super->s_dev)
 			holder = NULL;
-		journal->j_dev_bd = blkdev_get_by_dev(jdev, blkdev_mode, holder,
-						      NULL);
-		if (IS_ERR(journal->j_dev_bd)) {
-			result = PTR_ERR(journal->j_dev_bd);
-			journal->j_dev_bd = NULL;
+		journal->j_bdev_handle = bdev_open_by_dev(jdev, blkdev_mode,
+							  holder, NULL);
+		if (IS_ERR(journal->j_bdev_handle)) {
+			result = PTR_ERR(journal->j_bdev_handle);
+			journal->j_bdev_handle = NULL;
 			reiserfs_warning(super, "sh-458",
 					 "cannot init journal device unknown-block(%u,%u): %i",
 					 MAJOR(jdev), MINOR(jdev), result);
 			return result;
 		} else if (jdev != super->s_dev)
-			set_blocksize(journal->j_dev_bd, super->s_blocksize);
+			set_blocksize(journal->j_bdev_handle->bdev,
+				      super->s_blocksize);
 
 		return 0;
 	}
 
-	journal->j_dev_bd = blkdev_get_by_path(jdev_name, blkdev_mode, holder,
-					       NULL);
-	if (IS_ERR(journal->j_dev_bd)) {
-		result = PTR_ERR(journal->j_dev_bd);
-		journal->j_dev_bd = NULL;
+	journal->j_bdev_handle = bdev_open_by_path(jdev_name, blkdev_mode,
+						   holder, NULL);
+	if (IS_ERR(journal->j_bdev_handle)) {
+		result = PTR_ERR(journal->j_bdev_handle);
+		journal->j_bdev_handle = NULL;
 		reiserfs_warning(super, "sh-457",
 				 "journal_init_dev: Cannot open '%s': %i",
 				 jdev_name, result);
 		return result;
 	}
 
-	set_blocksize(journal->j_dev_bd, super->s_blocksize);
+	set_blocksize(journal->j_bdev_handle->bdev, super->s_blocksize);
 	reiserfs_info(super,
 		      "journal_init_dev: journal device: %pg\n",
-		      journal->j_dev_bd);
+		      journal->j_bdev_handle->bdev);
 	return 0;
 }
 
@@ -2810,7 +2804,7 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 				 "journal header magic %x (device %pg) does "
 				 "not match to magic found in super block %x",
 				 jh->jh_journal.jp_journal_magic,
-				 journal->j_dev_bd,
+				 journal->j_bdev_handle->bdev,
 				 sb_jp_journal_magic(rs));
 		brelse(bhjh);
 		goto free_and_return;
@@ -2834,7 +2828,7 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 	reiserfs_info(sb, "journal params: device %pg, size %u, "
 		      "journal first block %u, max trans len %u, max batch %u, "
 		      "max commit age %u, max trans age %u\n",
-		      journal->j_dev_bd,
+		      journal->j_bdev_handle->bdev,
 		      SB_ONDISK_JOURNAL_SIZE(sb),
 		      SB_ONDISK_JOURNAL_1st_BLOCK(sb),
 		      journal->j_trans_max,
fs/reiserfs/procfs.c +1 -1
index 3dba8acf4e83..83cb9402e0f9 100644
@@ -354,7 +354,7 @@ static int show_journal(struct seq_file *m, void *unused)
 		   "prepare: \t%12lu\n"
 		   "prepare_retry: \t%12lu\n",
 		   DJP(jp_journal_1st_block),
-		   SB_JOURNAL(sb)->j_dev_bd,
+		   SB_JOURNAL(sb)->j_bdev_handle->bdev,
 		   DJP(jp_journal_dev),
 		   DJP(jp_journal_size),
 		   DJP(jp_journal_trans_max),
fs/reiserfs/reiserfs.h +7 -4
index 55e85256aae8..8e426392b5c2 100644
@@ -299,7 +299,7 @@ struct reiserfs_journal {
 	/* oldest journal block.  start here for traverse */
 	struct reiserfs_journal_cnode *j_first;
 
-	struct block_device *j_dev_bd;
+	struct bdev_handle *j_bdev_handle;
 
 	/* first block on s_dev of reserved area journal */
 	int j_1st_reserved_block;
@@ -2809,9 +2809,12 @@ struct reiserfs_journal_header {
 #define journal_hash(t,sb,block) ((t)[_jhashfn((sb),(block)) & JBH_HASH_MASK])
 
 /* We need these to make journal.c code more readable */
-#define journal_find_get_block(s, block) __find_get_block(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize)
-#define journal_getblk(s, block) __getblk(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize)
-#define journal_bread(s, block) __bread(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize)
+#define journal_find_get_block(s, block) __find_get_block(\
+		SB_JOURNAL(s)->j_bdev_handle->bdev, block, s->s_blocksize)
+#define journal_getblk(s, block) __getblk(SB_JOURNAL(s)->j_bdev_handle->bdev,\
+		block, s->s_blocksize)
+#define journal_bread(s, block) __bread(SB_JOURNAL(s)->j_bdev_handle->bdev,\
+		block, s->s_blocksize)
 
 enum reiserfs_bh_state_bits {
 	BH_JDirty = BH_PrivateStart,	/* buffer is in current transaction */
Signature · 1 line
2.35.3
author reviewer
#94 lore raw

PATCH v3 27/29 reiserfs: Convert to bdev_open_by_dev/path()

In reply to Jan Kara

Convert reiserfs to use bdev_open_by_dev/path() and pass the handle around.

---
 fs/reiserfs/journal.c  | 56 +++++++++++++++++++-----------------------
 fs/reiserfs/procfs.c   |  2 +-
 fs/reiserfs/reiserfs.h | 11 ++++++---
 3 files changed, 33 insertions(+), 36 deletions(-)
fs/reiserfs/journal.c +25 -31
index 015bfe4e4524..171c912af50f 100644
@@ -90,8 +90,7 @@ static int flush_commit_list(struct super_block *s,
 static int can_dirty(struct reiserfs_journal_cnode *cn);
 static int journal_join(struct reiserfs_transaction_handle *th,
 			struct super_block *sb);
-static void release_journal_dev(struct super_block *super,
-			       struct reiserfs_journal *journal);
+static void release_journal_dev(struct reiserfs_journal *journal);
 static void dirty_one_transaction(struct super_block *s,
 				 struct reiserfs_journal_list *jl);
 static void flush_async_commits(struct work_struct *work);
@@ -1893,7 +1892,7 @@ static void free_journal_ram(struct super_block *sb)
 	 * j_header_bh is on the journal dev, make sure
 	 * not to release the journal dev until we brelse j_header_bh
 	 */
-	release_journal_dev(sb, journal);
+	release_journal_dev(journal);
 	vfree(journal);
 }
 
@@ -2387,7 +2386,7 @@ static int journal_read(struct super_block *sb)
 
 	cur_dblock = SB_ONDISK_JOURNAL_1st_BLOCK(sb);
 	reiserfs_info(sb, "checking transaction log (%pg)\n",
-		      journal->j_dev_bd);
+		      journal->j_bdev_handle->bdev);
 	start = ktime_get_seconds();
 
 	/*
@@ -2448,7 +2447,7 @@ static int journal_read(struct super_block *sb)
 		 * device and journal device to be the same
 		 */
 		d_bh =
-		    reiserfs_breada(journal->j_dev_bd, cur_dblock,
+		    reiserfs_breada(journal->j_bdev_handle->bdev, cur_dblock,
 				    sb->s_blocksize,
 				    SB_ONDISK_JOURNAL_1st_BLOCK(sb) +
 				    SB_ONDISK_JOURNAL_SIZE(sb));
@@ -2587,17 +2586,11 @@ static void journal_list_init(struct super_block *sb)
 	SB_JOURNAL(sb)->j_current_jl = alloc_journal_list(sb);
 }
 
-static void release_journal_dev(struct super_block *super,
-			       struct reiserfs_journal *journal)
+static void release_journal_dev(struct reiserfs_journal *journal)
 {
-	if (journal->j_dev_bd != NULL) {
-		void *holder = NULL;
-
-		if (journal->j_dev_bd->bd_dev != super->s_dev)
-			holder = journal;
-
-		blkdev_put(journal->j_dev_bd, holder);
-		journal->j_dev_bd = NULL;
+	if (journal->j_bdev_handle) {
+		bdev_release(journal->j_bdev_handle);
+		journal->j_bdev_handle = NULL;
 	}
 }
 
@@ -2612,7 +2605,7 @@ static int journal_init_dev(struct super_block *super,
 
 	result = 0;
 
-	journal->j_dev_bd = NULL;
+	journal->j_bdev_handle = NULL;
 	jdev = SB_ONDISK_JOURNAL_DEVICE(super) ?
 	    new_decode_dev(SB_ONDISK_JOURNAL_DEVICE(super)) : super->s_dev;
 
@@ -2623,36 +2616,37 @@ static int journal_init_dev(struct super_block *super,
 	if ((!jdev_name || !jdev_name[0])) {
 		if (jdev == super->s_dev)
 			holder = NULL;
-		journal->j_dev_bd = blkdev_get_by_dev(jdev, blkdev_mode, holder,
-						      NULL);
-		if (IS_ERR(journal->j_dev_bd)) {
-			result = PTR_ERR(journal->j_dev_bd);
-			journal->j_dev_bd = NULL;
+		journal->j_bdev_handle = bdev_open_by_dev(jdev, blkdev_mode,
+							  holder, NULL);
+		if (IS_ERR(journal->j_bdev_handle)) {
+			result = PTR_ERR(journal->j_bdev_handle);
+			journal->j_bdev_handle = NULL;
 			reiserfs_warning(super, "sh-458",
 					 "cannot init journal device unknown-block(%u,%u): %i",
 					 MAJOR(jdev), MINOR(jdev), result);
 			return result;
 		} else if (jdev != super->s_dev)
-			set_blocksize(journal->j_dev_bd, super->s_blocksize);
+			set_blocksize(journal->j_bdev_handle->bdev,
+				      super->s_blocksize);
 
 		return 0;
 	}
 
-	journal->j_dev_bd = blkdev_get_by_path(jdev_name, blkdev_mode, holder,
-					       NULL);
-	if (IS_ERR(journal->j_dev_bd)) {
-		result = PTR_ERR(journal->j_dev_bd);
-		journal->j_dev_bd = NULL;
+	journal->j_bdev_handle = bdev_open_by_path(jdev_name, blkdev_mode,
+						   holder, NULL);
+	if (IS_ERR(journal->j_bdev_handle)) {
+		result = PTR_ERR(journal->j_bdev_handle);
+		journal->j_bdev_handle = NULL;
 		reiserfs_warning(super, "sh-457",
 				 "journal_init_dev: Cannot open '%s': %i",
 				 jdev_name, result);
 		return result;
 	}
 
-	set_blocksize(journal->j_dev_bd, super->s_blocksize);
+	set_blocksize(journal->j_bdev_handle->bdev, super->s_blocksize);
 	reiserfs_info(super,
 		      "journal_init_dev: journal device: %pg\n",
-		      journal->j_dev_bd);
+		      journal->j_bdev_handle->bdev);
 	return 0;
 }
 
@@ -2810,7 +2804,7 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 				 "journal header magic %x (device %pg) does "
 				 "not match to magic found in super block %x",
 				 jh->jh_journal.jp_journal_magic,
-				 journal->j_dev_bd,
+				 journal->j_bdev_handle->bdev,
 				 sb_jp_journal_magic(rs));
 		brelse(bhjh);
 		goto free_and_return;
@@ -2834,7 +2828,7 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 	reiserfs_info(sb, "journal params: device %pg, size %u, "
 		      "journal first block %u, max trans len %u, max batch %u, "
 		      "max commit age %u, max trans age %u\n",
-		      journal->j_dev_bd,
+		      journal->j_bdev_handle->bdev,
 		      SB_ONDISK_JOURNAL_SIZE(sb),
 		      SB_ONDISK_JOURNAL_1st_BLOCK(sb),
 		      journal->j_trans_max,
fs/reiserfs/procfs.c +1 -1
index 3dba8acf4e83..83cb9402e0f9 100644
@@ -354,7 +354,7 @@ static int show_journal(struct seq_file *m, void *unused)
 		   "prepare: \t%12lu\n"
 		   "prepare_retry: \t%12lu\n",
 		   DJP(jp_journal_1st_block),
-		   SB_JOURNAL(sb)->j_dev_bd,
+		   SB_JOURNAL(sb)->j_bdev_handle->bdev,
 		   DJP(jp_journal_dev),
 		   DJP(jp_journal_size),
 		   DJP(jp_journal_trans_max),
fs/reiserfs/reiserfs.h +7 -4
index 7d12b8c5b2fa..8ad41271c256 100644
@@ -299,7 +299,7 @@ struct reiserfs_journal {
 	/* oldest journal block.  start here for traverse */
 	struct reiserfs_journal_cnode *j_first;
 
-	struct block_device *j_dev_bd;
+	struct bdev_handle *j_bdev_handle;
 
 	/* first block on s_dev of reserved area journal */
 	int j_1st_reserved_block;
@@ -2809,9 +2809,12 @@ struct reiserfs_journal_header {
 #define journal_hash(t,sb,block) ((t)[_jhashfn((sb),(block)) & JBH_HASH_MASK])
 
 /* We need these to make journal.c code more readable */
-#define journal_find_get_block(s, block) __find_get_block(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize)
-#define journal_getblk(s, block) __getblk(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize)
-#define journal_bread(s, block) __bread(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize)
+#define journal_find_get_block(s, block) __find_get_block(\
+		SB_JOURNAL(s)->j_bdev_handle->bdev, block, s->s_blocksize)
+#define journal_getblk(s, block) __getblk(SB_JOURNAL(s)->j_bdev_handle->bdev,\
+		block, s->s_blocksize)
+#define journal_bread(s, block) __bread(SB_JOURNAL(s)->j_bdev_handle->bdev,\
+		block, s->s_blocksize)
 
 enum reiserfs_bh_state_bits {
 	BH_JDirty = BH_PrivateStart,	/* buffer is in current transaction */
Signature · 1 line
2.35.3
author reviewer
#95 lore raw

PATCH v3 28/29 xfs: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

---
 fs/xfs/xfs_buf.c   | 22 +++++++++----------
 fs/xfs/xfs_buf.h   |  3 ++-
 fs/xfs/xfs_super.c | 54 +++++++++++++++++++++++++---------------------
 3 files changed, 42 insertions(+), 37 deletions(-)
fs/xfs/xfs_buf.c +10 -12
index 3b903f6bce98..496e2f5fdcc6 100644
@@ -1938,8 +1938,6 @@ void
 xfs_free_buftarg(
 	struct xfs_buftarg	*btp)
 {
-	struct block_device	*bdev = btp->bt_bdev;
-
 	unregister_shrinker(&btp->bt_shrinker);
 	ASSERT(percpu_counter_sum(&btp->bt_io_count) == 0);
 	percpu_counter_destroy(&btp->bt_io_count);
@@ -1947,8 +1945,8 @@ xfs_free_buftarg(
 
 	fs_put_dax(btp->bt_daxdev, btp->bt_mount);
 	/* the main block device is closed by kill_block_super */
-	if (bdev != btp->bt_mount->m_super->s_bdev)
-		blkdev_put(bdev, btp->bt_mount->m_super);
+	if (btp->bt_bdev != btp->bt_mount->m_super->s_bdev)
+		bdev_release(btp->bt_bdev_handle);
 
 	kmem_free(btp);
 }
@@ -1983,16 +1981,15 @@ xfs_setsize_buftarg(
  */
 STATIC int
 xfs_setsize_buftarg_early(
-	xfs_buftarg_t		*btp,
-	struct block_device	*bdev)
+	xfs_buftarg_t		*btp)
 {
-	return xfs_setsize_buftarg(btp, bdev_logical_block_size(bdev));
+	return xfs_setsize_buftarg(btp, bdev_logical_block_size(btp->bt_bdev));
 }
 
 struct xfs_buftarg *
 xfs_alloc_buftarg(
 	struct xfs_mount	*mp,
-	struct block_device	*bdev)
+	struct bdev_handle	*bdev_handle)
 {
 	xfs_buftarg_t		*btp;
 	const struct dax_holder_operations *ops = NULL;
@@ -2003,9 +2000,10 @@ xfs_alloc_buftarg(
 	btp = kmem_zalloc(sizeof(*btp), KM_NOFS);
 
 	btp->bt_mount = mp;
-	btp->bt_dev =  bdev->bd_dev;
-	btp->bt_bdev = bdev;
-	btp->bt_daxdev = fs_dax_get_by_bdev(bdev, &btp->bt_dax_part_off,
+	btp->bt_bdev_handle = bdev_handle;
+	btp->bt_dev =  bdev_handle->bdev->bd_dev;
+	btp->bt_bdev = bdev_handle->bdev;
+	btp->bt_daxdev = fs_dax_get_by_bdev(btp->bt_bdev, &btp->bt_dax_part_off,
 					    mp, ops);
 
 	/*
@@ -2015,7 +2013,7 @@ xfs_alloc_buftarg(
 	ratelimit_state_init(&btp->bt_ioerror_rl, 30 * HZ,
 			     DEFAULT_RATELIMIT_BURST);
 
-	if (xfs_setsize_buftarg_early(btp, bdev))
+	if (xfs_setsize_buftarg_early(btp))
 		goto error_free;
 
 	if (list_lru_init(&btp->bt_lru))
fs/xfs/xfs_buf.h +2 -1
index 549c60942208..f6418c1312f5 100644
@@ -92,6 +92,7 @@ typedef unsigned int xfs_buf_flags_t;
  */
 typedef struct xfs_buftarg {
 	dev_t			bt_dev;
+	struct bdev_handle	*bt_bdev_handle;
 	struct block_device	*bt_bdev;
 	struct dax_device	*bt_daxdev;
 	u64			bt_dax_part_off;
@@ -351,7 +352,7 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset)
  *	Handling of buftargs.
  */
 struct xfs_buftarg *xfs_alloc_buftarg(struct xfs_mount *mp,
-		struct block_device *bdev);
+		struct bdev_handle *bdev_handle);
 extern void xfs_free_buftarg(struct xfs_buftarg *);
 extern void xfs_buftarg_wait(struct xfs_buftarg *);
 extern void xfs_buftarg_drain(struct xfs_buftarg *);
fs/xfs/xfs_super.c +30 -24

116 lines not on this page · raw message

Signature · 1 line
2.35.3
#96 lore raw

Re: [PATCH 28/29] xfs: Convert to bdev_open_by_path()

Quoted 1 line · Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

....

Quoted 10 lines
fs/xfs/xfs_super.c +2 -2
@@ -426,15 +427,15 @@ xfs_shutdown_devices(
 	 * race, everyone loses.
 	 */
 	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
-		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_logdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_logdev_targp->bt_bdev_handle->bdev);
 	}
 	if (mp->m_rtdev_targp) {

fs/xfs/xfs_super.c xfs_shutdown_devices( · line 439

Quoted 10 lines
fs/xfs/xfs_super.c +4 -4
-		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_rtdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_rtdev_targp->bt_bdev_handle->bdev);
 	}
-	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev);
-	invalidate_bdev(mp->m_ddev_targp->bt_bdev);
+	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev_handle->bdev);
+	invalidate_bdev(mp->m_ddev_targp->bt_bdev_handle->bdev);
 }

Why do these need to be converted to run through bt_bdev_handle? If the buftarg is present and we've assigned targp->bt_bdev_handle during the call to xfs_alloc_buftarg(), then we've assigned targp->bt_bdev from the handle at the same time, yes?

Cheers,

Dave.

Signature · 2 lines
Dave Chinner
[email protected]
author reviewer
#97 lore raw

Re: [PATCH 28/29] xfs: Convert to bdev_open_by_path()

Quoted 21 lines · Dave Chinner
Quoted 1 line · Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

....

Quoted 17 lines
fs/xfs/xfs_super.c +4 -6
@@ -426,15 +427,15 @@ xfs_shutdown_devices(
 	 * race, everyone loses.
 	 */
 	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
-		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_logdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_logdev_targp->bt_bdev_handle->bdev);
 	}
 	if (mp->m_rtdev_targp) {
-		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_rtdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_rtdev_targp->bt_bdev_handle->bdev);
 	}
-	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev);
-	invalidate_bdev(mp->m_ddev_targp->bt_bdev);
Quoted 8 lines

fs/xfs/xfs_super.c xfs_shutdown_devices( · line 439

Quoted 3 lines
fs/xfs/xfs_super.c +2 -0
+	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev_handle->bdev);
+	invalidate_bdev(mp->m_ddev_targp->bt_bdev_handle->bdev);
 }

Why do these need to be converted to run through bt_bdev_handle? If the buftarg is present and we've assigned targp->bt_bdev_handle during the call to xfs_alloc_buftarg(), then we've assigned targp->bt_bdev from the handle at the same time, yes?

Good point, these conversions are pointless now so I've removed them. They are leftover from a previous version based on a kernel where xfs was dropping bdev references in a different place. Thanks for noticing.

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
author reviewer Acked-by Christoph Hellwig
#98 lore raw

Re: [PATCH 28/29] xfs: Convert to bdev_open_by_path()

Quoted 31 lines · Jan Kara
Quoted 21 lines · Dave Chinner
Quoted 1 line · Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

....

Quoted 17 lines
(quoted diff) +4 -6
@@ -426,15 +427,15 @@ xfs_shutdown_devices(
 	 * race, everyone loses.
 	 */
 	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
-		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_logdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_logdev_targp->bt_bdev_handle->bdev);
 	}
 	if (mp->m_rtdev_targp) {
-		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_rtdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_rtdev_targp->bt_bdev_handle->bdev);
 	}
-	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev);
-	invalidate_bdev(mp->m_ddev_targp->bt_bdev);
Quoted 8 lines

Review xfs_shutdown_devices( · line 439

Quoted 3 lines
(quoted diff) +2 -0
+	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev_handle->bdev);
+	invalidate_bdev(mp->m_ddev_targp->bt_bdev_handle->bdev);
 }

Why do these need to be converted to run through bt_bdev_handle? If the buftarg is present and we've assigned targp->bt_bdev_handle during the call to xfs_alloc_buftarg(), then we've assigned targp->bt_bdev from the handle at the same time, yes?

Quoted 3 lines

Good point, these conversions are pointless now so I've removed them. They are leftover from a previous version based on a kernel where xfs was dropping bdev references in a different place. Thanks for noticing.

FWIW attached is a new version of the patch I have currently in my tree.

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR

From 7ba12a848a8dcf70029c897b8cf7bca6d4f8c098 Mon Sep 17 00:00:00 2001
From: Jan Kara <[email protected]>
Date: Fri, 23 Jun 2023 15:48:53 +0200
Subject: [PATCH] xfs: Convert to bdev_open_by_path()

Convert xfs to use bdev_open_by_path() and pass the handle around.

---
 fs/xfs/xfs_buf.c   | 22 ++++++++++------------
 fs/xfs/xfs_buf.h   |  3 ++-
 fs/xfs/xfs_super.c | 42 ++++++++++++++++++++++++------------------
 3 files changed, 36 insertions(+), 31 deletions(-)
fs/xfs/xfs_buf.c +10 -12
index 3b903f6bce98..75a98eec2b27 100644
@@ -1938,8 +1938,6 @@ void
 xfs_free_buftarg(
 	struct xfs_buftarg	*btp)
 {
-	struct block_device	*bdev = btp->bt_bdev;
-
 	unregister_shrinker(&btp->bt_shrinker);
 	ASSERT(percpu_counter_sum(&btp->bt_io_count) == 0);
 	percpu_counter_destroy(&btp->bt_io_count);
@@ -1947,8 +1945,8 @@ xfs_free_buftarg(
 
 	fs_put_dax(btp->bt_daxdev, btp->bt_mount);
 	/* the main block device is closed by kill_block_super */
-	if (bdev != btp->bt_mount->m_super->s_bdev)
-		blkdev_put(bdev, btp->bt_mount->m_super);
+	if (btp->bt_bdev != btp->bt_mount->m_super->s_bdev)
+		bdev_release(btp->bt_bdev_handle);
 
 	kmem_free(btp);
 }
@@ -1983,16 +1981,15 @@ xfs_setsize_buftarg(
  */
 STATIC int
 xfs_setsize_buftarg_early(
-	xfs_buftarg_t		*btp,
-	struct block_device	*bdev)
+	xfs_buftarg_t		*btp)
 {
-	return xfs_setsize_buftarg(btp, bdev_logical_block_size(bdev));
+	return xfs_setsize_buftarg(btp, bdev_logical_block_size(btp->bt_bdev));
 }
 
 struct xfs_buftarg *
 xfs_alloc_buftarg(
 	struct xfs_mount	*mp,
-	struct block_device	*bdev)
+	struct bdev_handle	*bdev_handle)
 {
 	xfs_buftarg_t		*btp;
 	const struct dax_holder_operations *ops = NULL;
@@ -2003,9 +2000,10 @@ xfs_alloc_buftarg(
 	btp = kmem_zalloc(sizeof(*btp), KM_NOFS);
 
 	btp->bt_mount = mp;
-	btp->bt_dev =  bdev->bd_dev;
-	btp->bt_bdev = bdev;
-	btp->bt_daxdev = fs_dax_get_by_bdev(bdev, &btp->bt_dax_part_off,
+	btp->bt_bdev_handle = bdev_handle;
+	btp->bt_dev = bdev_handle->bdev->bd_dev;
+	btp->bt_bdev = bdev_handle->bdev;
+	btp->bt_daxdev = fs_dax_get_by_bdev(btp->bt_bdev, &btp->bt_dax_part_off,
 					    mp, ops);
 
 	/*
@@ -2015,7 +2013,7 @@ xfs_alloc_buftarg(
 	ratelimit_state_init(&btp->bt_ioerror_rl, 30 * HZ,
 			     DEFAULT_RATELIMIT_BURST);
 
-	if (xfs_setsize_buftarg_early(btp, bdev))
+	if (xfs_setsize_buftarg_early(btp))
 		goto error_free;
 
 	if (list_lru_init(&btp->bt_lru))
fs/xfs/xfs_buf.h +2 -1
index 549c60942208..f6418c1312f5 100644
@@ -92,6 +92,7 @@ typedef unsigned int xfs_buf_flags_t;
  */
 typedef struct xfs_buftarg {
 	dev_t			bt_dev;
+	struct bdev_handle	*bt_bdev_handle;
 	struct block_device	*bt_bdev;
 	struct dax_device	*bt_daxdev;
 	u64			bt_dax_part_off;
@@ -351,7 +352,7 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset)
  *	Handling of buftargs.
  */
 struct xfs_buftarg *xfs_alloc_buftarg(struct xfs_mount *mp,
-		struct block_device *bdev);
+		struct bdev_handle *bdev_handle);
 extern void xfs_free_buftarg(struct xfs_buftarg *);
 extern void xfs_buftarg_wait(struct xfs_buftarg *);
 extern void xfs_buftarg_drain(struct xfs_buftarg *);
fs/xfs/xfs_super.c +24 -18

95 lines not on this page · raw message

Signature · 1 line
2.35.3
Reviewed-by
#99 lore raw

Re: [PATCH 28/29] xfs: Convert to bdev_open_by_path()

Quoted 32 lines · Jan Kara
Quoted 31 lines · Jan Kara
Quoted 29 lines · Dave Chinner
Quoted 1 line · Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

....

Quoted 10 lines
(quoted diff) +2 -2
@@ -426,15 +427,15 @@ xfs_shutdown_devices(
 	 * race, everyone loses.
 	 */
 	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
-		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_logdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_logdev_targp->bt_bdev_handle->bdev);
 	}
 	if (mp->m_rtdev_targp) {

Review xfs_shutdown_devices( · line 439

Quoted 10 lines
(quoted diff) +4 -4
-		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_rtdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_rtdev_targp->bt_bdev_handle->bdev);
 	}
-	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev);
-	invalidate_bdev(mp->m_ddev_targp->bt_bdev);
+	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev_handle->bdev);
+	invalidate_bdev(mp->m_ddev_targp->bt_bdev_handle->bdev);
 }

Why do these need to be converted to run through bt_bdev_handle? If the buftarg is present and we've assigned targp->bt_bdev_handle during the call to xfs_alloc_buftarg(), then we've assigned targp->bt_bdev from the handle at the same time, yes?

Quoted 5 lines
Quoted 3 lines

Good point, these conversions are pointless now so I've removed them. They are leftover from a previous version based on a kernel where xfs was dropping bdev references in a different place. Thanks for noticing.

FWIW attached is a new version of the patch I have currently in my tree.

Looks OK.

Signature · 2 lines
Dave Chinner
[email protected]
maintainer Reviewed-by
#100 lore raw

Re: [PATCH 28/29] xfs: Convert to bdev_open_by_path()

In reply to Jan Kara
Quoted 32 lines · Jan Kara
Quoted 31 lines · Jan Kara
Quoted 29 lines · Dave Chinner
Quoted 1 line · Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

....

Quoted 10 lines
(quoted diff) +2 -2
@@ -426,15 +427,15 @@ xfs_shutdown_devices(
 	 * race, everyone loses.
 	 */
 	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
-		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_logdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_logdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_logdev_targp->bt_bdev_handle->bdev);
 	}
 	if (mp->m_rtdev_targp) {

Review xfs_shutdown_devices( · line 439

Quoted 10 lines
(quoted diff) +4 -4
-		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev);
-		invalidate_bdev(mp->m_rtdev_targp->bt_bdev);
+		blkdev_issue_flush(mp->m_rtdev_targp->bt_bdev_handle->bdev);
+		invalidate_bdev(mp->m_rtdev_targp->bt_bdev_handle->bdev);
 	}
-	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev);
-	invalidate_bdev(mp->m_ddev_targp->bt_bdev);
+	blkdev_issue_flush(mp->m_ddev_targp->bt_bdev_handle->bdev);
+	invalidate_bdev(mp->m_ddev_targp->bt_bdev_handle->bdev);
 }

Why do these need to be converted to run through bt_bdev_handle? If the buftarg is present and we've assigned targp->bt_bdev_handle during the call to xfs_alloc_buftarg(), then we've assigned targp->bt_bdev from the handle at the same time, yes?

Quoted 5 lines
Quoted 3 lines

Good point, these conversions are pointless now so I've removed them. They are leftover from a previous version based on a kernel where xfs was dropping bdev references in a different place. Thanks for noticing.

FWIW attached is a new version of the patch I have currently in my tree.

The appended patch looks good to me,

author reviewer
#101 lore raw

PATCH v3 28/29 xfs: Convert to bdev_open_by_path()

In reply to Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

---
 fs/xfs/xfs_buf.c   | 22 ++++++++++------------
 fs/xfs/xfs_buf.h   |  3 ++-
 fs/xfs/xfs_super.c | 42 ++++++++++++++++++++++++------------------
 3 files changed, 36 insertions(+), 31 deletions(-)
fs/xfs/xfs_buf.c +10 -12
index c1ece4a08ff4..003e157241da 100644
@@ -1945,8 +1945,6 @@ void
 xfs_free_buftarg(
 	struct xfs_buftarg	*btp)
 {
-	struct block_device	*bdev = btp->bt_bdev;
-
 	unregister_shrinker(&btp->bt_shrinker);
 	ASSERT(percpu_counter_sum(&btp->bt_io_count) == 0);
 	percpu_counter_destroy(&btp->bt_io_count);
@@ -1954,8 +1952,8 @@ xfs_free_buftarg(
 
 	fs_put_dax(btp->bt_daxdev, btp->bt_mount);
 	/* the main block device is closed by kill_block_super */
-	if (bdev != btp->bt_mount->m_super->s_bdev)
-		blkdev_put(bdev, btp->bt_mount->m_super);
+	if (btp->bt_bdev != btp->bt_mount->m_super->s_bdev)
+		bdev_release(btp->bt_bdev_handle);
 
 	kmem_free(btp);
 }
@@ -1990,16 +1988,15 @@ xfs_setsize_buftarg(
  */
 STATIC int
 xfs_setsize_buftarg_early(
-	xfs_buftarg_t		*btp,
-	struct block_device	*bdev)
+	xfs_buftarg_t		*btp)
 {
-	return xfs_setsize_buftarg(btp, bdev_logical_block_size(bdev));
+	return xfs_setsize_buftarg(btp, bdev_logical_block_size(btp->bt_bdev));
 }
 
 struct xfs_buftarg *
 xfs_alloc_buftarg(
 	struct xfs_mount	*mp,
-	struct block_device	*bdev)
+	struct bdev_handle	*bdev_handle)
 {
 	xfs_buftarg_t		*btp;
 	const struct dax_holder_operations *ops = NULL;
@@ -2010,9 +2007,10 @@ xfs_alloc_buftarg(
 	btp = kmem_zalloc(sizeof(*btp), KM_NOFS);
 
 	btp->bt_mount = mp;
-	btp->bt_dev =  bdev->bd_dev;
-	btp->bt_bdev = bdev;
-	btp->bt_daxdev = fs_dax_get_by_bdev(bdev, &btp->bt_dax_part_off,
+	btp->bt_bdev_handle = bdev_handle;
+	btp->bt_dev = bdev_handle->bdev->bd_dev;
+	btp->bt_bdev = bdev_handle->bdev;
+	btp->bt_daxdev = fs_dax_get_by_bdev(btp->bt_bdev, &btp->bt_dax_part_off,
 					    mp, ops);
 
 	/*
@@ -2022,7 +2020,7 @@ xfs_alloc_buftarg(
 	ratelimit_state_init(&btp->bt_ioerror_rl, 30 * HZ,
 			     DEFAULT_RATELIMIT_BURST);
 
-	if (xfs_setsize_buftarg_early(btp, bdev))
+	if (xfs_setsize_buftarg_early(btp))
 		goto error_free;
 
 	if (list_lru_init(&btp->bt_lru))
fs/xfs/xfs_buf.h +2 -1
index df8f47953bb4..ada9d310b7d3 100644
@@ -98,6 +98,7 @@ typedef unsigned int xfs_buf_flags_t;
  */
 typedef struct xfs_buftarg {
 	dev_t			bt_dev;
+	struct bdev_handle	*bt_bdev_handle;
 	struct block_device	*bt_bdev;
 	struct dax_device	*bt_daxdev;
 	u64			bt_dax_part_off;
@@ -364,7 +365,7 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset)
  *	Handling of buftargs.
  */
 struct xfs_buftarg *xfs_alloc_buftarg(struct xfs_mount *mp,
-		struct block_device *bdev);
+		struct bdev_handle *bdev_handle);
 extern void xfs_free_buftarg(struct xfs_buftarg *);
 extern void xfs_buftarg_wait(struct xfs_buftarg *);
 extern void xfs_buftarg_drain(struct xfs_buftarg *);
fs/xfs/xfs_super.c +24 -18

95 lines not on this page · raw message

Signature · 1 line
2.35.3
Acked-by Reviewed-by on v3 28/29
#102 lore raw

Re: [PATCH 28/29] xfs: Convert to bdev_open_by_path()

Quoted 8 lines · Jan Kara

Convert xfs to use bdev_open_by_path() and pass the handle around.

Looks like a pretty straightforward conversion, so:

For the block parts,

For this patch:

--D

Quoted 201 lines
---
 fs/xfs/xfs_buf.c   | 22 ++++++++++------------
 fs/xfs/xfs_buf.h   |  3 ++-
 fs/xfs/xfs_super.c | 42 ++++++++++++++++++++++++------------------
 3 files changed, 36 insertions(+), 31 deletions(-)
fs/xfs/xfs_buf.c +10 -12
index c1ece4a08ff4..003e157241da 100644
@@ -1945,8 +1945,6 @@ void
 xfs_free_buftarg(
 	struct xfs_buftarg	*btp)
 {
-	struct block_device	*bdev = btp->bt_bdev;
-
 	unregister_shrinker(&btp->bt_shrinker);
 	ASSERT(percpu_counter_sum(&btp->bt_io_count) == 0);
 	percpu_counter_destroy(&btp->bt_io_count);
@@ -1954,8 +1952,8 @@ xfs_free_buftarg(
 
 	fs_put_dax(btp->bt_daxdev, btp->bt_mount);
 	/* the main block device is closed by kill_block_super */
-	if (bdev != btp->bt_mount->m_super->s_bdev)
-		blkdev_put(bdev, btp->bt_mount->m_super);
+	if (btp->bt_bdev != btp->bt_mount->m_super->s_bdev)
+		bdev_release(btp->bt_bdev_handle);
 
 	kmem_free(btp);
 }
@@ -1990,16 +1988,15 @@ xfs_setsize_buftarg(
  */
 STATIC int
 xfs_setsize_buftarg_early(
-	xfs_buftarg_t		*btp,
-	struct block_device	*bdev)
+	xfs_buftarg_t		*btp)
 {
-	return xfs_setsize_buftarg(btp, bdev_logical_block_size(bdev));
+	return xfs_setsize_buftarg(btp, bdev_logical_block_size(btp->bt_bdev));
 }
 
 struct xfs_buftarg *
 xfs_alloc_buftarg(
 	struct xfs_mount	*mp,
-	struct block_device	*bdev)
+	struct bdev_handle	*bdev_handle)
 {
 	xfs_buftarg_t		*btp;
 	const struct dax_holder_operations *ops = NULL;
@@ -2010,9 +2007,10 @@ xfs_alloc_buftarg(
 	btp = kmem_zalloc(sizeof(*btp), KM_NOFS);
 
 	btp->bt_mount = mp;
-	btp->bt_dev =  bdev->bd_dev;
-	btp->bt_bdev = bdev;
-	btp->bt_daxdev = fs_dax_get_by_bdev(bdev, &btp->bt_dax_part_off,
+	btp->bt_bdev_handle = bdev_handle;
+	btp->bt_dev = bdev_handle->bdev->bd_dev;
+	btp->bt_bdev = bdev_handle->bdev;
+	btp->bt_daxdev = fs_dax_get_by_bdev(btp->bt_bdev, &btp->bt_dax_part_off,
 					    mp, ops);
 
 	/*
@@ -2022,7 +2020,7 @@ xfs_alloc_buftarg(
 	ratelimit_state_init(&btp->bt_ioerror_rl, 30 * HZ,
 			     DEFAULT_RATELIMIT_BURST);
 
-	if (xfs_setsize_buftarg_early(btp, bdev))
+	if (xfs_setsize_buftarg_early(btp))
 		goto error_free;
 
 	if (list_lru_init(&btp->bt_lru))
fs/xfs/xfs_buf.h +2 -1
index df8f47953bb4..ada9d310b7d3 100644
@@ -98,6 +98,7 @@ typedef unsigned int xfs_buf_flags_t;
  */
 typedef struct xfs_buftarg {
 	dev_t			bt_dev;
+	struct bdev_handle	*bt_bdev_handle;
 	struct block_device	*bt_bdev;
 	struct dax_device	*bt_daxdev;
 	u64			bt_dax_part_off;
@@ -364,7 +365,7 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset)
  *	Handling of buftargs.
  */
 struct xfs_buftarg *xfs_alloc_buftarg(struct xfs_mount *mp,
-		struct block_device *bdev);
+		struct bdev_handle *bdev_handle);
 extern void xfs_free_buftarg(struct xfs_buftarg *);
 extern void xfs_buftarg_wait(struct xfs_buftarg *);
 extern void xfs_buftarg_drain(struct xfs_buftarg *);
fs/xfs/xfs_super.c +24 -18
index 819a3568b28f..f0ae07828153 100644
@@ -361,14 +361,15 @@ STATIC int
 xfs_blkdev_get(
 	xfs_mount_t		*mp,
 	const char		*name,
-	struct block_device	**bdevp)
+	struct bdev_handle	**handlep)
 {
 	int			error = 0;
 
-	*bdevp = blkdev_get_by_path(name, BLK_OPEN_READ | BLK_OPEN_WRITE,
-				    mp->m_super, &fs_holder_ops);
-	if (IS_ERR(*bdevp)) {
-		error = PTR_ERR(*bdevp);
+	*handlep = bdev_open_by_path(name, BLK_OPEN_READ | BLK_OPEN_WRITE,
+				     mp->m_super, &fs_holder_ops);
+	if (IS_ERR(*handlep)) {
+		error = PTR_ERR(*handlep);
+		*handlep = NULL;
 		xfs_warn(mp, "Invalid device [%s], error=%d", name, error);
 	}
 
@@ -433,7 +434,7 @@ xfs_open_devices(
 {
 	struct super_block	*sb = mp->m_super;
 	struct block_device	*ddev = sb->s_bdev;
-	struct block_device	*logdev = NULL, *rtdev = NULL;
+	struct bdev_handle	*logdev_handle = NULL, *rtdev_handle = NULL;
 	int			error;
 
 	/*
@@ -446,17 +447,19 @@ xfs_open_devices(
 	 * Open real time and log devices - order is important.
 	 */
 	if (mp->m_logname) {
-		error = xfs_blkdev_get(mp, mp->m_logname, &logdev);
+		error = xfs_blkdev_get(mp, mp->m_logname, &logdev_handle);
 		if (error)
 			goto out_relock;
 	}
 
 	if (mp->m_rtname) {
-		error = xfs_blkdev_get(mp, mp->m_rtname, &rtdev);
+		error = xfs_blkdev_get(mp, mp->m_rtname, &rtdev_handle);
 		if (error)
 			goto out_close_logdev;
 
-		if (rtdev == ddev || rtdev == logdev) {
+		if (rtdev_handle->bdev == ddev ||
+		    (logdev_handle &&
+		     rtdev_handle->bdev == logdev_handle->bdev)) {
 			xfs_warn(mp,
 	"Cannot mount filesystem with identical rtdev and ddev/logdev.");
 			error = -EINVAL;
@@ -468,22 +471,25 @@ xfs_open_devices(
 	 * Setup xfs_mount buffer target pointers
 	 */
 	error = -ENOMEM;
-	mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev);
+	mp->m_ddev_targp = xfs_alloc_buftarg(mp, sb->s_bdev_handle);
 	if (!mp->m_ddev_targp)
 		goto out_close_rtdev;
 
-	if (rtdev) {
-		mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev);
+	if (rtdev_handle) {
+		mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev_handle);
 		if (!mp->m_rtdev_targp)
 			goto out_free_ddev_targ;
 	}
 
-	if (logdev && logdev != ddev) {
-		mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev);
+	if (logdev_handle && logdev_handle->bdev != ddev) {
+		mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev_handle);
 		if (!mp->m_logdev_targp)
 			goto out_free_rtdev_targ;
 	} else {
 		mp->m_logdev_targp = mp->m_ddev_targp;
+		/* Handle won't be used, drop it */
+		if (logdev_handle)
+			bdev_release(logdev_handle);
 	}
 
 	error = 0;
@@ -497,11 +503,11 @@ xfs_open_devices(
  out_free_ddev_targ:
 	xfs_free_buftarg(mp->m_ddev_targp);
  out_close_rtdev:
-	 if (rtdev)
-		 blkdev_put(rtdev, sb);
+	 if (rtdev_handle)
+		bdev_release(rtdev_handle);
  out_close_logdev:
-	if (logdev && logdev != ddev)
-		blkdev_put(logdev, sb);
+	if (logdev_handle)
+		bdev_release(logdev_handle);
 	goto out_relock;
 }
 
Signature · 1 line
2.35.3
author reviewer
#103 lore raw

PATCH v3 29/29 block: Remove blkdev_get_by_*() functions

In reply to Jan Kara

blkdev_get_by_*() and blkdev_put() functions are now unused. Remove them.

---
 block/bdev.c           | 94 ++++++++++++++----------------------------
 include/linux/blkdev.h |  5 ---
 2 files changed, 30 insertions(+), 69 deletions(-)
block/bdev.c +30 -64
index f1de1e65517b..5624cac0aea8 100644
@@ -729,7 +729,7 @@ void blkdev_put_no_open(struct block_device *bdev)
 }
 	
 /**
- * blkdev_get_by_dev - open a block device by device number
+ * bdev_open_by_dev - open a block device by device number
  * @dev: device number of block device to open
  * @mode: open mode (BLK_OPEN_*)
  * @holder: exclusive holder identifier
@@ -741,32 +741,40 @@ void blkdev_put_no_open(struct block_device *bdev)
  *
  * Use this interface ONLY if you really do not have anything better - i.e. when
  * you are behind a truly sucky interface and all you are given is a device
- * number.  Everything else should use blkdev_get_by_path().
+ * number.  Everything else should use bdev_open_by_path().
  *
  * CONTEXT:
  * Might sleep.
  *
  * RETURNS:
- * Reference to the block_device on success, ERR_PTR(-errno) on failure.
+ * Handle with a reference to the block_device on success, ERR_PTR(-errno) on
+ * failure.
  */
-struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
-		const struct blk_holder_ops *hops)
+struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
+				     const struct blk_holder_ops *hops)
 {
-	bool unblock_events = true;
+	struct bdev_handle *handle = kmalloc(sizeof(struct bdev_handle),
+					     GFP_KERNEL);
 	struct block_device *bdev;
+	bool unblock_events = true;
 	struct gendisk *disk;
 	int ret;
 
+	if (!handle)
+		return ERR_PTR(-ENOMEM);
+
 	ret = devcgroup_check_permission(DEVCG_DEV_BLOCK,
 			MAJOR(dev), MINOR(dev),
 			((mode & BLK_OPEN_READ) ? DEVCG_ACC_READ : 0) |
 			((mode & BLK_OPEN_WRITE) ? DEVCG_ACC_WRITE : 0));
 	if (ret)
-		return ERR_PTR(ret);
+		goto free_handle;
 
 	bdev = blkdev_get_no_open(dev);
-	if (!bdev)
-		return ERR_PTR(-ENXIO);
+	if (!bdev) {
+		ret = -ENXIO;
+		goto free_handle;
+	}
 	disk = bdev->bd_disk;
 
 	if (holder) {
@@ -815,7 +823,10 @@ struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 
 	if (unblock_events)
 		disk_unblock_events(disk);
-	return bdev;
+	handle->bdev = bdev;
+	handle->holder = holder;
+	handle->mode = mode;
+	return handle;
 put_module:
 	module_put(disk->fops->owner);
 abort_claiming:
@@ -825,34 +836,14 @@ struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 	disk_unblock_events(disk);
 put_blkdev:
 	blkdev_put_no_open(bdev);
+free_handle:
+	kfree(handle);
 	return ERR_PTR(ret);
 }
-EXPORT_SYMBOL(blkdev_get_by_dev);
-
-struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
-				     const struct blk_holder_ops *hops)
-{
-	struct bdev_handle *handle = kmalloc(sizeof(*handle), GFP_KERNEL);
-	struct block_device *bdev;
-
-	if (!handle)
-		return ERR_PTR(-ENOMEM);
-	bdev = blkdev_get_by_dev(dev, mode, holder, hops);
-	if (IS_ERR(bdev)) {
-		kfree(handle);
-		return ERR_CAST(bdev);
-	}
-	handle->bdev = bdev;
-	handle->holder = holder;
-	if (holder)
-		mode |= BLK_OPEN_EXCL;
-	handle->mode = mode;
-	return handle;
-}
 EXPORT_SYMBOL(bdev_open_by_dev);
 
 /**
- * blkdev_get_by_path - open a block device by name
+ * bdev_open_by_path - open a block device by name
  * @path: path to the block device to open
  * @mode: open mode (BLK_OPEN_*)
  * @holder: exclusive holder identifier
@@ -866,29 +857,9 @@ EXPORT_SYMBOL(bdev_open_by_dev);
  * Might sleep.
  *
  * RETURNS:
- * Reference to the block_device on success, ERR_PTR(-errno) on failure.
+ * Handle with a reference to the block_device on success, ERR_PTR(-errno) on
+ * failure.
  */
-struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
-		void *holder, const struct blk_holder_ops *hops)
-{
-	struct block_device *bdev;
-	dev_t dev;
-	int error;
-
-	error = lookup_bdev(path, &dev);
-	if (error)
-		return ERR_PTR(error);
-
-	bdev = blkdev_get_by_dev(dev, mode, holder, hops);
-	if (!IS_ERR(bdev) && (mode & BLK_OPEN_WRITE) && bdev_read_only(bdev)) {
-		blkdev_put(bdev, holder);
-		return ERR_PTR(-EACCES);
-	}
-
-	return bdev;
-}
-EXPORT_SYMBOL(blkdev_get_by_path);
-
 struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
 		void *holder, const struct blk_holder_ops *hops)
 {
@@ -911,8 +882,9 @@ struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
 }
 EXPORT_SYMBOL(bdev_open_by_path);
 
-void blkdev_put(struct block_device *bdev, void *holder)
+void bdev_release(struct bdev_handle *handle)
 {
+	struct block_device *bdev = handle->bdev;
 	struct gendisk *disk = bdev->bd_disk;
 
 	/*
@@ -926,8 +898,8 @@ void blkdev_put(struct block_device *bdev, void *holder)
 		sync_blockdev(bdev);
 
 	mutex_lock(&disk->open_mutex);
-	if (holder)
-		bd_end_claim(bdev, holder);
+	if (handle->holder)
+		bd_end_claim(bdev, handle->holder);
 
 	/*
 	 * Trigger event checking and tell drivers to flush MEDIA_CHANGE
@@ -944,12 +916,6 @@ void blkdev_put(struct block_device *bdev, void *holder)
 
 	module_put(disk->fops->owner);
 	blkdev_put_no_open(bdev);
-}
-EXPORT_SYMBOL(blkdev_put);
-
-void bdev_release(struct bdev_handle *handle)
-{
-	blkdev_put(handle->bdev, handle->holder);
 	kfree(handle);
 }
 EXPORT_SYMBOL(bdev_release);
include/linux/blkdev.h +0 -5
index ae741dec184b..c54981ecda44 100644
@@ -1484,10 +1484,6 @@ struct bdev_handle {
 	blk_mode_t mode;
 };
 
-struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
-		const struct blk_holder_ops *hops);
-struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
-		void *holder, const struct blk_holder_ops *hops);
 struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 		const struct blk_holder_ops *hops);
 struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
@@ -1495,7 +1491,6 @@ struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
 int bd_prepare_to_claim(struct block_device *bdev, void *holder,
 		const struct blk_holder_ops *hops);
 void bd_abort_claiming(struct block_device *bdev, void *holder);
-void blkdev_put(struct block_device *bdev, void *holder);
 void bdev_release(struct bdev_handle *handle);
 
 /* just for blk-cgroup, don't use elsewhere */
Signature · 1 line
2.35.3
author reviewer
#105 lore raw

PATCH v3 29/29 block: Remove blkdev_get_by_*() functions

In reply to Jan Kara

blkdev_get_by_*() and blkdev_put() functions are now unused. Remove them.

---
 block/bdev.c           | 94 ++++++++++++++----------------------------
 include/linux/blkdev.h |  5 ---
 2 files changed, 30 insertions(+), 69 deletions(-)
block/bdev.c +30 -64
index 4628dcb1da8a..ea80958027a8 100644
@@ -729,7 +729,7 @@ void blkdev_put_no_open(struct block_device *bdev)
 }
 	
 /**
- * blkdev_get_by_dev - open a block device by device number
+ * bdev_open_by_dev - open a block device by device number
  * @dev: device number of block device to open
  * @mode: open mode (BLK_OPEN_*)
  * @holder: exclusive holder identifier
@@ -741,32 +741,40 @@ void blkdev_put_no_open(struct block_device *bdev)
  *
  * Use this interface ONLY if you really do not have anything better - i.e. when
  * you are behind a truly sucky interface and all you are given is a device
- * number.  Everything else should use blkdev_get_by_path().
+ * number.  Everything else should use bdev_open_by_path().
  *
  * CONTEXT:
  * Might sleep.
  *
  * RETURNS:
- * Reference to the block_device on success, ERR_PTR(-errno) on failure.
+ * Handle with a reference to the block_device on success, ERR_PTR(-errno) on
+ * failure.
  */
-struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
-		const struct blk_holder_ops *hops)
+struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
+				     const struct blk_holder_ops *hops)
 {
-	bool unblock_events = true;
+	struct bdev_handle *handle = kmalloc(sizeof(struct bdev_handle),
+					     GFP_KERNEL);
 	struct block_device *bdev;
+	bool unblock_events = true;
 	struct gendisk *disk;
 	int ret;
 
+	if (!handle)
+		return ERR_PTR(-ENOMEM);
+
 	ret = devcgroup_check_permission(DEVCG_DEV_BLOCK,
 			MAJOR(dev), MINOR(dev),
 			((mode & BLK_OPEN_READ) ? DEVCG_ACC_READ : 0) |
 			((mode & BLK_OPEN_WRITE) ? DEVCG_ACC_WRITE : 0));
 	if (ret)
-		return ERR_PTR(ret);
+		goto free_handle;
 
 	bdev = blkdev_get_no_open(dev);
-	if (!bdev)
-		return ERR_PTR(-ENXIO);
+	if (!bdev) {
+		ret = -ENXIO;
+		goto free_handle;
+	}
 	disk = bdev->bd_disk;
 
 	if (holder) {
@@ -815,7 +823,10 @@ struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 
 	if (unblock_events)
 		disk_unblock_events(disk);
-	return bdev;
+	handle->bdev = bdev;
+	handle->holder = holder;
+	handle->mode = mode;
+	return handle;
 put_module:
 	module_put(disk->fops->owner);
 abort_claiming:
@@ -825,34 +836,14 @@ struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 	disk_unblock_events(disk);
 put_blkdev:
 	blkdev_put_no_open(bdev);
+free_handle:
+	kfree(handle);
 	return ERR_PTR(ret);
 }
-EXPORT_SYMBOL(blkdev_get_by_dev);
-
-struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
-				     const struct blk_holder_ops *hops)
-{
-	struct bdev_handle *handle = kmalloc(sizeof(*handle), GFP_KERNEL);
-	struct block_device *bdev;
-
-	if (!handle)
-		return ERR_PTR(-ENOMEM);
-	bdev = blkdev_get_by_dev(dev, mode, holder, hops);
-	if (IS_ERR(bdev)) {
-		kfree(handle);
-		return ERR_CAST(bdev);
-	}
-	handle->bdev = bdev;
-	handle->holder = holder;
-	if (holder)
-		mode |= BLK_OPEN_EXCL;
-	handle->mode = mode;
-	return handle;
-}
 EXPORT_SYMBOL(bdev_open_by_dev);
 
 /**
- * blkdev_get_by_path - open a block device by name
+ * bdev_open_by_path - open a block device by name
  * @path: path to the block device to open
  * @mode: open mode (BLK_OPEN_*)
  * @holder: exclusive holder identifier
@@ -866,29 +857,9 @@ EXPORT_SYMBOL(bdev_open_by_dev);
  * Might sleep.
  *
  * RETURNS:
- * Reference to the block_device on success, ERR_PTR(-errno) on failure.
+ * Handle with a reference to the block_device on success, ERR_PTR(-errno) on
+ * failure.
  */
-struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
-		void *holder, const struct blk_holder_ops *hops)
-{
-	struct block_device *bdev;
-	dev_t dev;
-	int error;
-
-	error = lookup_bdev(path, &dev);
-	if (error)
-		return ERR_PTR(error);
-
-	bdev = blkdev_get_by_dev(dev, mode, holder, hops);
-	if (!IS_ERR(bdev) && (mode & BLK_OPEN_WRITE) && bdev_read_only(bdev)) {
-		blkdev_put(bdev, holder);
-		return ERR_PTR(-EACCES);
-	}
-
-	return bdev;
-}
-EXPORT_SYMBOL(blkdev_get_by_path);
-
 struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
 		void *holder, const struct blk_holder_ops *hops)
 {
@@ -911,8 +882,9 @@ struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
 }
 EXPORT_SYMBOL(bdev_open_by_path);
 
-void blkdev_put(struct block_device *bdev, void *holder)
+void bdev_release(struct bdev_handle *handle)
 {
+	struct block_device *bdev = handle->bdev;
 	struct gendisk *disk = bdev->bd_disk;
 
 	/*
@@ -926,8 +898,8 @@ void blkdev_put(struct block_device *bdev, void *holder)
 		sync_blockdev(bdev);
 
 	mutex_lock(&disk->open_mutex);
-	if (holder)
-		bd_end_claim(bdev, holder);
+	if (handle->holder)
+		bd_end_claim(bdev, handle->holder);
 
 	/*
 	 * Trigger event checking and tell drivers to flush MEDIA_CHANGE
@@ -944,12 +916,6 @@ void blkdev_put(struct block_device *bdev, void *holder)
 
 	module_put(disk->fops->owner);
 	blkdev_put_no_open(bdev);
-}
-EXPORT_SYMBOL(blkdev_put);
-
-void bdev_release(struct bdev_handle *handle)
-{
-	blkdev_put(handle->bdev, handle->holder);
 	kfree(handle);
 }
 EXPORT_SYMBOL(bdev_release);
include/linux/blkdev.h +0 -5
index 51fa7ffdee83..98406eea5d1b 100644
@@ -1485,10 +1485,6 @@ struct bdev_handle {
 	blk_mode_t mode;
 };
 
-struct block_device *blkdev_get_by_dev(dev_t dev, blk_mode_t mode, void *holder,
-		const struct blk_holder_ops *hops);
-struct block_device *blkdev_get_by_path(const char *path, blk_mode_t mode,
-		void *holder, const struct blk_holder_ops *hops);
 struct bdev_handle *bdev_open_by_dev(dev_t dev, blk_mode_t mode, void *holder,
 		const struct blk_holder_ops *hops);
 struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
@@ -1496,7 +1492,6 @@ struct bdev_handle *bdev_open_by_path(const char *path, blk_mode_t mode,
 int bd_prepare_to_claim(struct block_device *bdev, void *holder,
 		const struct blk_holder_ops *hops);
 void bd_abort_claiming(struct block_device *bdev, void *holder);
-void blkdev_put(struct block_device *bdev, void *holder);
 void bdev_release(struct bdev_handle *handle);
 
 /* just for blk-cgroup, don't use elsewhere */
Signature · 1 line
2.35.3
maintainer Acked-by on the v3 series
#106 lore raw
In reply to Jan Kara
Quoted 4 lines · On Wed, Sep 27, 2023 at 3:34?AM Jan Kara <[email protected]>…

Hello,

this is a v3 of the patch series which implements the idea of blkdev_get_by_*()

v4?

Quoted 13 lines

calls returning bdev_handle which is then passed to blkdev_put() [1]. This makes the get and put calls for bdevs more obviously matching and allows us to propagate context from get to put without having to modify all the users (again!). In particular I need to propagate used open flags to blkdev_put() to be able count writeable opens and add support for blocking writes to mounted block devices. I'll send that series separately.

The series is based on Btrfs tree's for-next branch [2] as of today as the series depends on Christoph's changes to btrfs device handling. Patches have passed some reasonable testing - I've tested block changes, md, dm, bcache, xfs, btrfs, ext4, swap. More testing or review is always welcome. Thanks! I've pushed out the full branch to:

Quoted 4 lines

git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs.git bdev_handle

to ease review / testing. Christian, can you pull the patches to your tree to get some exposure in linux-next as well? Thanks!

For the block bits:

Signature · 1 line
Jens Axboe
maintainer Applied
#107 lore raw
In reply to Jan Kara
Quoted 11 lines · Jan Kara

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. This will eventually allow us to pass one more argument to blkdev_put() (renamed to bdev_release()) without too much hassle.

[...]

to ease review / testing. Christian, can you pull the patches to your tree to get some exposure in linux-next as well? Thanks!

Yep. So I did it slighly differently. I pulled in the btrfs prereqs and then applied your series on top of it so we get all the Link: tags right. I'm running tests right now. Please double-check.

---

Applied to the vfs.super branch of the vfs/vfs.git tree. Patches in the vfs.super branch should appear in linux-next soon.

Please report any outstanding bugs that were missed during review in a new review to the original patch series allowing us to drop it.

It's encouraged to provide Acked-bys and Reviewed-bys even though the patch has now been applied. If possible patch trailers will be updated.

Note that commit hashes shown below are subject to change due to rebase, trailer updates or similar. If in doubt, please check the listed branch.

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git
branch: vfs.super

[01/29] block: Provide bdev_open_* functions

       https://git.kernel.org/vfs/vfs/c/b7c828aa0b3c
[02/29] block: Use bdev_open_by_dev() in blkdev_open()
        https://git.kernel.org/vfs/vfs/c/d4e36f27b45a
[03/29] block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset()
        https://git.kernel.org/vfs/vfs/c/5f9bd6764c7a
[04/29] drdb: Convert to use bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/0220ca8e443d
[05/29] pktcdvd: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/7af10b889789
[06/29] rnbd-srv: Convert to use bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/3d27892a4be7
[07/29] xen/blkback: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/26afb0ed10b3
[08/29] zram: Convert to use bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/efc8e3f4c6dc
[09/29] bcache: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/dc893f51d24a
[10/29] dm: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/80c2267c6d07
[11/29] md: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/15db36126ca6
[12/29] mtd: block2mtd: Convert to bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/4c27234bf3ce
[13/29] nvmet: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/70cffddcc300
[14/29] s390/dasd: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/5581d03457f8
[15/29] scsi: target: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/43de7d844d47
[16/29] PM: hibernate: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/105ea4a2fd18
[17/29] PM: hibernate: Drop unused snapshot_test argument
        https://git.kernel.org/vfs/vfs/c/b589a66e3688
[18/29] mm/swap: Convert to use bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/615af8e29233
[19/29] fs: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/5173192bcfe6
[20/29] btrfs: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/8cf64782764f
[21/29] erofs: Convert to use bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/4d41880bf249
[22/29] ext4: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/f7507612395e
[23/29] f2fs: Convert to bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/d9ff8e3b6498
[24/29] jfs: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/459dc6376338
[25/29] nfs/blocklayout: Convert to use bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/5b1df9a40929
[26/29] ocfs2: Convert to use bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/b6b95acbd943
[27/29] reiserfs: Convert to bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/7e3615ff6119
[28/29] xfs: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/176ccb99e207
[29/29] block: Remove blkdev_get_by_*() functions
        https://git.kernel.org/vfs/vfs/c/953863a5a2ff
author reviewer
#108 lore raw
Quoted 8 lines · Christian Brauner
Quoted 7 lines · Jan Kara

Create struct bdev_handle that contains all parameters that need to be passed to blkdev_put() and provide bdev_open_* functions that return this structure instead of plain bdev pointer. This will eventually allow us to pass one more argument to blkdev_put() (renamed to bdev_release()) without too much hassle.

Quoted 8 lines
Quoted 4 lines

[...]

to ease review / testing. Christian, can you pull the patches to your tree to get some exposure in linux-next as well? Thanks!

Yep. So I did it slighly differently. I pulled in the btrfs prereqs and then applied your series on top of it so we get all the Link: tags right. I'm running tests right now. Please double-check.

Thanks for picking patches up! I've checked the branch and it looks good to me.

Honza

Quoted 76 lines

---

Applied to the vfs.super branch of the vfs/vfs.git tree. Patches in the vfs.super branch should appear in linux-next soon.

Please report any outstanding bugs that were missed during review in a new review to the original patch series allowing us to drop it.

It's encouraged to provide Acked-bys and Reviewed-bys even though the patch has now been applied. If possible patch trailers will be updated.

Note that commit hashes shown below are subject to change due to rebase, trailer updates or similar. If in doubt, please check the listed branch.

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git
branch: vfs.super

[01/29] block: Provide bdev_open_* functions

       https://git.kernel.org/vfs/vfs/c/b7c828aa0b3c
[02/29] block: Use bdev_open_by_dev() in blkdev_open()
        https://git.kernel.org/vfs/vfs/c/d4e36f27b45a
[03/29] block: Use bdev_open_by_dev() in disk_scan_partitions() and blkdev_bszset()
        https://git.kernel.org/vfs/vfs/c/5f9bd6764c7a
[04/29] drdb: Convert to use bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/0220ca8e443d
[05/29] pktcdvd: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/7af10b889789
[06/29] rnbd-srv: Convert to use bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/3d27892a4be7
[07/29] xen/blkback: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/26afb0ed10b3
[08/29] zram: Convert to use bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/efc8e3f4c6dc
[09/29] bcache: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/dc893f51d24a
[10/29] dm: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/80c2267c6d07
[11/29] md: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/15db36126ca6
[12/29] mtd: block2mtd: Convert to bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/4c27234bf3ce
[13/29] nvmet: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/70cffddcc300
[14/29] s390/dasd: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/5581d03457f8
[15/29] scsi: target: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/43de7d844d47
[16/29] PM: hibernate: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/105ea4a2fd18
[17/29] PM: hibernate: Drop unused snapshot_test argument
        https://git.kernel.org/vfs/vfs/c/b589a66e3688
[18/29] mm/swap: Convert to use bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/615af8e29233
[19/29] fs: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/5173192bcfe6
[20/29] btrfs: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/8cf64782764f
[21/29] erofs: Convert to use bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/4d41880bf249
[22/29] ext4: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/f7507612395e
[23/29] f2fs: Convert to bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/d9ff8e3b6498
[24/29] jfs: Convert to bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/459dc6376338
[25/29] nfs/blocklayout: Convert to use bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/5b1df9a40929
[26/29] ocfs2: Convert to use bdev_open_by_dev()
        https://git.kernel.org/vfs/vfs/c/b6b95acbd943
[27/29] reiserfs: Convert to bdev_open_by_dev/path()
        https://git.kernel.org/vfs/vfs/c/7e3615ff6119
[28/29] xfs: Convert to bdev_open_by_path()
        https://git.kernel.org/vfs/vfs/c/176ccb99e207
[29/29] block: Remove blkdev_get_by_*() functions
        https://git.kernel.org/vfs/vfs/c/953863a5a2ff
Signature · 2 lines
Jan Kara <jack-IBi9RG/[email protected]>
SUSE Labs, CR