Copyright (c) 2013 Sylvestre Gallon <ccna.syl@gmail.com>
Copyright (c) 2018 Helg Bredow <helg@openbsd.org>
Permission to use, copy, modify, and distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
.Dd $Mdocdate: March 12 2021 $ .Dt FUSE_NEW 3 .Os .Sh NAME .Nm fuse_new .Nd FUSE implementation routine to initialise the FUSE connection .Sh SYNOPSIS n fuse.h .Ft struct fuse * .Fn fuse_new "struct fuse_chan *fc" "struct fuse_args *args" \ "const struct fuse_operations *ops" "unused size_t size" \ "void *userdata" .Sh DESCRIPTION Initialises the FUSE library on the channel returned by .Xr fuse_mount 3 .
p FUSE operations work in the same way as their UNIX file system counterparts. A major exception is that these routines return a negated errno value (-errno) on failure.
p All operations are optional but a functional file system will want to implement at least statfs, readdir, open, read and getattr. FUSE will return ENOSYS if any operation other than flush, fsync or fsyncdir is not implemented.
p The first parameter to each of these operations (except for init and terminate) is a NULL terminated string representing the full path to the file or directory, relative to the root of this file system, that is being operated on. d -literal struct fuse_operations { int (*getattr)(const char *, struct stat *); int (*readlink)(const char *, char *, size_t); int (*getdir)(const char *, fuse_dirh_t, fuse_dirfil_t); int (*mknod)(const char *, mode_t, dev_t); int (*mkdir)(const char *, mode_t); int (*unlink)(const char *); int (*rmdir)(const char *); int (*symlink)(const char *, const char *); int (*rename)(const char *, const char *); int (*link)(const char *, const char *); int (*chmod)(const char *, mode_t); int (*chown)(const char *, uid_t, gid_t); int (*truncate)(const char *, off_t); int (*utime)(const char *, struct utimbuf *); int (*open)(const char *, struct fuse_file_info *); int (*read)(const char *, char *, size_t, off_t, struct fuse_file_info *); int (*write)(const char *, const char *, size_t, off_t, struct fuse_file_info *); int (*statfs)(const char *, struct statvfs *); int (*flush)(const char *, struct fuse_file_info *); int (*release)(const char *, struct fuse_file_info *); int (*fsync)(const char *, int, struct fuse_file_info *); int (*setxattr)(const char *, const char *, const char *, size_t int); int (*getxattr)(const char *, const char *, char *, size_t); int (*listxattr)(const char *, char *, size_t); int (*removexattr)(const char *, const char *); int (*opendir)(const char *, struct fuse_file_info *); int (*readdir)(const char *, void *, fuse_fill_dir_t, off_t, struct fuse_file_info *); int (*releasedir)(const char *, struct fuse_file_info *); int (*fsyncdir)(const char *, int, struct fuse_file_info *); void *(*init)(struct fuse_conn_info *); void (*destroy)(void *); int (*access)(const char *, int); int (*create)(const char *, mode_t, struct fuse_file_info *); int (*ftruncate)(const char *, off_t, struct fuse_file_info *); int (*fgetattr)(const char *, struct stat *, struct fuse_file_info *); int (*lock)(const char *, struct fuse_file_info *, int, struct flock *); int (*utimens)(const char *, const struct timespec *); int (*bmap)(const char *, size_t , uint64_t *); }; .Ed
p The order of calls is: d -literal -offset indent init ... opendir readdir releasedir open read write ... flush release ... destroy .Ed l -tag -width "releasedir" t access Not implemented. .Ox always behaves as if the default_permissions mount option was specified. See .Xr fuse_mount 3 . t chmod Called when file access permissions are changed. t chown Called when either the file owner or group is changed. t create Not implemented on .Ox . File systems must implement mknod instead. In the reference implementation this is an atomic operation that both creates and opens the file. There is no equivalent in the .Ox VFS. t flush Called when the file is closed by the .Xr close 2 system call. This is the only way for a file system to return an error on close. t fsync Optional function that implements .Xr fsync 2 and .Xr fdatasync 2 . The datasync parameter specifies whether the operation is as a result of a call to .Xr fdatasync 2 and is currently always 0 (false). ffi.fh_id will contain the file handle returned by the file system when the file was opened. t fsyncdir Not implemented. t getattr Corresponds to the .Xr stat 2 system call. Flags and extended attributes are ignored. This operation is mandatory. t getxattr Not implemented. t getdir Deprecated. File system should implement readdir instead. t mknod Called on .Xr open 2 and .Xr mknod 2 to create regular files, pipes and device special files. t open Called on .Xr open 2 . Due to the difference between FUSE and the .Ox VFS, open will only be called once for each file for every different combination of flags provided to .Xr open 2 . The O_CREAT and O_TRUNC flags are never passed from the kernel to open, the mknod and truncate operations are invoked before open instead. t opendir Called when a directory is opened for reading. t release Called when there are no more references to the file. t releasedir Called when there are no more references to the directory. t setattr Equivalent to .Xr chown 2 and .Xr chmod 2 system calls. Setting file flags is not supported. t setxattr Not implemented. .El
p Options supported by args are: l -tag -width "readdir_ino" t debug, -d Print debug information to stdout. t gid=%u The GID that will be reported as the group for all files by getattr. t hard_remove Immediately delete a file even if it's currently open by a process. Otherwise FUSE will temporarily rename the file and only delete it when it is no longer referenced. This is to avoid the file system having to deal with this situation. This is always set on .Ox . t readdir_ino Similar to use_ino but the file system's inode number is only reported for readdir. This is always set on .Ox because it's required by .Xr getcwd 3 . t uid=%u The UID that will be reported as the owner for all files by getattr. t umask=%o The file mode mask applied to the permission for all files by getattr. t use_ino By default, FUSE will return an internal inode number for getattr and readdir and this will be different every time the file system is mounted. If this is set, the file system's own inode number will be reported instead. Useful only for file system that have inode numbers. .El .Sh SEE ALSO .Xr fuse_get_context 3 , .Xr fuse_main 3 , .Xr fuse_mount 3 .Sh STANDARDS The .Fn fuse_new function conforms to FUSE 2.6. .Sh HISTORY The .Fn fuse_new function first appeared in .Ox 5.4 . .Sh AUTHORS .An Sylvestre Gallon Aq Mt ccna.syl@gmail.com .An Helg Bredow Aq Mt helg@openbsd.org