From 29732938a6289a15e907da234d6692a2ead71855 Mon Sep 17 00:00:00 2001 From: Zach Brown Date: Tue, 10 Nov 2015 16:53:30 -0500 Subject: vfs: add copy_file_range syscall and vfs helper Add a copy_file_range() system call for offloading copies between regular files. This gives an interface to underlying layers of the storage stack which can copy without reading and writing all the data. There are a few candidates that should support copy offloading in the nearer term: - btrfs shares extent references with its clone ioctl - NFS has patches to add a COPY command which copies on the server - SCSI has a family of XCOPY commands which copy in the device This system call avoids the complexity of also accelerating the creation of the destination file by operating on an existing destination file descriptor, not a path. Currently the high level vfs entry point limits copy offloading to files on the same mount and super (and not in the same file). This can be relaxed if we get implementations which can copy between file systems safely. Signed-off-by: Zach Brown [Anna Schumaker: Change -EINVAL to -EBADF during file verification, Change flags parameter from int to unsigned int, Add function to include/linux/syscalls.h, Check copy len after file open mode, Don't forbid ranges inside the same file, Use rw_verify_area() to veriy ranges, Use file_out rather than file_in, Add COPY_FR_REFLINK flag] Signed-off-by: Anna Schumaker Reviewed-by: Christoph Hellwig Signed-off-by: Al Viro --- include/linux/fs.h | 3 +++ include/linux/syscalls.h | 3 +++ 2 files changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 3aa514254161..e8a736242b1a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1629,6 +1629,7 @@ struct file_operations { #ifndef CONFIG_MMU unsigned (*mmap_capabilities)(struct file *); #endif + ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, loff_t, size_t, unsigned int); }; struct inode_operations { @@ -1680,6 +1681,8 @@ extern ssize_t vfs_readv(struct file *, const struct iovec __user *, unsigned long, loff_t *); extern ssize_t vfs_writev(struct file *, const struct iovec __user *, unsigned long, loff_t *); +extern ssize_t vfs_copy_file_range(struct file *, loff_t , struct file *, + loff_t, size_t, unsigned int); struct super_operations { struct inode *(*alloc_inode)(struct super_block *sb); diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index c2b66a277e98..185815c96433 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -886,6 +886,9 @@ asmlinkage long sys_execveat(int dfd, const char __user *filename, const char __user *const __user *envp, int flags); asmlinkage long sys_membarrier(int cmd, int flags); +asmlinkage long sys_copy_file_range(int fd_in, loff_t __user *off_in, + int fd_out, loff_t __user *off_out, + size_t len, unsigned int flags); asmlinkage long sys_mlock2(unsigned long start, size_t len, int flags); -- cgit v1.2.3 From acc15575e78e534c12549d8057a692f490a50f61 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 3 Dec 2015 12:59:49 +0100 Subject: locks: new locks_mandatory_area calling convention Pass a loff_t end for the last byte instead of the 32-bit count parameter to allow full file clones even on 32-bit architectures. While we're at it also simplify the read/write selection. Signed-off-by: Christoph Hellwig Acked-by: J. Bruce Fields Signed-off-by: Al Viro --- include/linux/fs.h | 30 ++++++++++++++---------------- 1 file changed, 14 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index e8a736242b1a..4377b2df991d 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2030,12 +2030,9 @@ extern struct kobject *fs_kobj; #define MAX_RW_COUNT (INT_MAX & PAGE_CACHE_MASK) -#define FLOCK_VERIFY_READ 1 -#define FLOCK_VERIFY_WRITE 2 - #ifdef CONFIG_FILE_LOCKING extern int locks_mandatory_locked(struct file *); -extern int locks_mandatory_area(int, struct inode *, struct file *, loff_t, size_t); +extern int locks_mandatory_area(struct inode *, struct file *, loff_t, loff_t, unsigned char); /* * Candidates for mandatory locking have the setgid bit set @@ -2065,17 +2062,19 @@ static inline int locks_verify_locked(struct file *file) } static inline int locks_verify_truncate(struct inode *inode, - struct file *filp, + struct file *f, loff_t size) { - if (inode->i_flctx && mandatory_lock(inode)) - return locks_mandatory_area( - FLOCK_VERIFY_WRITE, inode, filp, - size < inode->i_size ? size : inode->i_size, - (size < inode->i_size ? inode->i_size - size - : size - inode->i_size) - ); - return 0; + if (!inode->i_flctx || !mandatory_lock(inode)) + return 0; + + if (size < inode->i_size) { + return locks_mandatory_area(inode, f, size, inode->i_size - 1, + F_WRLCK); + } else { + return locks_mandatory_area(inode, f, inode->i_size, size - 1, + F_WRLCK); + } } static inline int break_lease(struct inode *inode, unsigned int mode) @@ -2144,9 +2143,8 @@ static inline int locks_mandatory_locked(struct file *file) return 0; } -static inline int locks_mandatory_area(int rw, struct inode *inode, - struct file *filp, loff_t offset, - size_t count) +static inline int locks_mandatory_area(struct inode *inode, struct file *filp, + loff_t start, loff_t end, unsigned char type) { return 0; } -- cgit v1.2.3 From 04b38d601239b4d9be641b412cf4b7456a041c67 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 3 Dec 2015 12:59:50 +0100 Subject: vfs: pull btrfs clone API to vfs layer The btrfs clone ioctls are now adopted by other file systems, with NFS and CIFS already having support for them, and XFS being under active development. To avoid growth of various slightly incompatible implementations, add one to the VFS. Note that clones are different from file copies in several ways: - they are atomic vs other writers - they support whole file clones - they support 64-bit legth clones - they do not allow partial success (aka short writes) - clones are expected to be a fast metadata operation Because of that it would be rather cumbersome to try to piggyback them on top of the recent clone_file_range infrastructure. The converse isn't true and the clone_file_range system call could try clone file range as a first attempt to copy, something that further patches will enable. Based on earlier work from Peng Tao. Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- include/linux/fs.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 4377b2df991d..5d987aefcf1e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1629,7 +1629,10 @@ struct file_operations { #ifndef CONFIG_MMU unsigned (*mmap_capabilities)(struct file *); #endif - ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, loff_t, size_t, unsigned int); + ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, + loff_t, size_t, unsigned int); + int (*clone_file_range)(struct file *, loff_t, struct file *, loff_t, + u64); }; struct inode_operations { @@ -1683,6 +1686,8 @@ extern ssize_t vfs_writev(struct file *, const struct iovec __user *, unsigned long, loff_t *); extern ssize_t vfs_copy_file_range(struct file *, loff_t , struct file *, loff_t, size_t, unsigned int); +extern int vfs_clone_file_range(struct file *file_in, loff_t pos_in, + struct file *file_out, loff_t pos_out, u64 len); struct super_operations { struct inode *(*alloc_inode)(struct super_block *sb); -- cgit v1.2.3 From ffa0160a103917defd5d9c097ae0455a59166e03 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 3 Dec 2015 12:59:52 +0100 Subject: nfsd: implement the NFSv4.2 CLONE operation This is basically a remote version of the btrfs CLONE operation, so the implementation is fairly trivial. Made even more trivial by stealing the XDR code and general framework Anna Schumaker's COPY prototype. Signed-off-by: Christoph Hellwig Acked-by: J. Bruce Fields Signed-off-by: Al Viro --- include/linux/nfs4.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index e7e78537aea2..43aeabd4b968 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -139,10 +139,10 @@ enum nfs_opnum4 { Needs to be updated if more operations are defined in future.*/ #define FIRST_NFS4_OP OP_ACCESS -#define LAST_NFS4_OP OP_WRITE_SAME #define LAST_NFS40_OP OP_RELEASE_LOCKOWNER #define LAST_NFS41_OP OP_RECLAIM_COMPLETE -#define LAST_NFS42_OP OP_WRITE_SAME +#define LAST_NFS42_OP OP_CLONE +#define LAST_NFS4_OP LAST_NFS42_OP enum nfsstat4 { NFS4_OK = 0, -- cgit v1.2.3 From 54dbc15172375641ef03399e8f911d7165eb90fb Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Sat, 19 Dec 2015 00:55:59 -0800 Subject: vfs: hoist the btrfs deduplication ioctl to the vfs Hoist the btrfs EXTENT_SAME ioctl up to the VFS and make the name more systematic (FIDEDUPERANGE). Signed-off-by: Darrick J. Wong Signed-off-by: Al Viro --- include/linux/fs.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 5d987aefcf1e..d71814b81a3c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1633,6 +1633,8 @@ struct file_operations { loff_t, size_t, unsigned int); int (*clone_file_range)(struct file *, loff_t, struct file *, loff_t, u64); + ssize_t (*dedupe_file_range)(struct file *, u64, u64, struct file *, + u64); }; struct inode_operations { @@ -1688,6 +1690,8 @@ extern ssize_t vfs_copy_file_range(struct file *, loff_t , struct file *, loff_t, size_t, unsigned int); extern int vfs_clone_file_range(struct file *file_in, loff_t pos_in, struct file *file_out, loff_t pos_out, u64 len); +extern int vfs_dedupe_file_range(struct file *file, + struct file_dedupe_range *same); struct super_operations { struct inode *(*alloc_inode)(struct super_block *sb); -- cgit v1.2.3