#include <lunaix/ds/btrie.h>
#include <lunaix/ds/hashtable.h>
#include <lunaix/ds/hstr.h>
+#include <lunaix/ds/ldga.h>
#include <lunaix/ds/llist.h>
#include <lunaix/ds/lru.h>
#include <lunaix/ds/mutex.h>
-#include <lunaix/process.h>
#include <lunaix/status.h>
-#include <stdatomic.h>
+#include <lunaix/spike.h>
+#include <lunaix/bcache.h>
+#include <lunaix/fs_acl.h>
+
+#include <usr/lunaix/fstypes.h>
#define VFS_NAME_MAXLEN 128
#define VFS_MAX_FD 32
-#define VFS_IFDIR 0x1
-#define VFS_IFFILE 0x2
-#define VFS_IFSEQDEV 0x4
-#define VFS_IFVOLDEV 0x8
-#define VFS_IFSYMLINK 0x16
+#define VFS_IFFILE F_FILE
+#define VFS_IFDIR (F_FILE | F_DIR )
+#define VFS_IFDEV (F_FILE | F_DEV )
+#define VFS_IFSYMLINK (F_FILE | F_SYMLINK)
+#define VFS_IFVOLDEV (F_FILE | F_SVDEV )
+#define VFS_IFSEQDEV VFS_IFDEV
+// Walk, mkdir if component encountered is non-exists.
#define VFS_WALK_MKPARENT 0x1
+
+// Walk, relative to current FS.
#define VFS_WALK_FSRELATIVE 0x2
+
+/*
+ Terminate the walk on the immediate parent,
+ name of child (last component) is returned through `component`
+*/
#define VFS_WALK_PARENT 0x4
+
+// Do not follow the symbolic link
#define VFS_WALK_NOFOLLOW 0x8
#define VFS_HASHTABLE_BITS 10
#define VFS_PATH_DELIM '/'
-#define FSTYPE_ROFS 0x1
-
-#define DO_STATUS(errno) SYSCALL_ESTATUS(__current->k_status = errno)
-#define DO_STATUS_OR_RETURN(errno) ({ errno < 0 ? DO_STATUS(errno) : errno; })
+#define FSTYPE_ROFS 0b00000001
+#define FSTYPE_PSEUDO 0x00000010
#define TEST_FD(fd) (fd >= 0 && fd < VFS_MAX_FD)
+#define EXPORT_FILE_SYSTEM(fs_id, init_fn) \
+ export_ldga_el(fs, fs_id, ptr_t, init_fn)
+
#define VFS_VALID_CHAR(chr) \
(('A' <= (chr) && (chr) <= 'Z') || ('a' <= (chr) && (chr) <= 'z') || \
('0' <= (chr) && (chr) <= '9') || (chr) == '.' || (chr) == '_' || \
- (chr) == '-')
+ (chr) == '-' || (chr) == ':')
-#define unlock_inode(inode) mutex_unlock(&inode->lock)
+#define unlock_inode(inode) mutex_unlock_nested(&inode->lock)
#define lock_inode(inode) \
({ \
- mutex_lock(&inode->lock); \
+ mutex_lock_nested(&inode->lock); \
lru_use_one(inode_lru, &inode->lru); \
})
-#define unlock_dnode(dnode) mutex_unlock(&dnode->lock)
+#define unlock_dnode(dnode) mutex_unlock_nested(&dnode->lock)
#define lock_dnode(dnode) \
({ \
- mutex_lock(&dnode->lock); \
+ mutex_lock_nested(&dnode->lock); \
lru_use_one(dnode_lru, &dnode->lru); \
})
-typedef uint32_t inode_t;
+#define assert_fs(cond) assert_p(cond, "FS")
+#define fail_fs(msg) fail_p(msg, "FS")
+
+typedef u32_t inode_t;
struct v_dnode;
struct v_inode;
extern struct hstr vfs_dot;
extern struct v_dnode* vfs_sysroot;
+typedef int (*mntops_mnt)(struct v_superblock* vsb, struct v_dnode* mount_point);
+typedef int (*mntops_umnt)(struct v_superblock* vsb);
+
struct filesystem
{
+ struct llist_header fs_flat;
struct hlist_node fs_list;
struct hstr fs_name;
- uint32_t types;
+ u32_t types;
int fs_id;
- int (*mount)(struct v_superblock* vsb, struct v_dnode* mount_point);
- int (*unmount)(struct v_superblock* vsb);
+ mntops_mnt mount;
+ mntops_umnt unmount;
+};
+
+struct fs_iter
+{
+ struct filesystem* fs;
};
struct v_superblock
struct device* dev;
struct v_dnode* root;
struct filesystem* fs;
- uint32_t iobuf_size;
+ struct blkbuf_cache* blks;
struct hbucket* i_cache;
+ struct hbucket* d_cache;
+
+ void* data;
+ unsigned int ref_count;
+ size_t blksize;
struct
{
- uint32_t (*read_capacity)(struct v_superblock* vsb);
- uint32_t (*read_usage)(struct v_superblock* vsb);
+ size_t (*read_capacity)(struct v_superblock* vsb);
+ size_t (*read_usage)(struct v_superblock* vsb);
void (*init_inode)(struct v_superblock* vsb, struct v_inode* inode);
+ void (*release)(struct v_superblock* vsb);
} ops;
};
struct dir_context
{
- int index;
void* cb_data;
void (*read_complete_callback)(struct dir_context* dctx,
const char* name,
{
int (*write)(struct v_inode* inode, void* buffer, size_t len, size_t fpos);
int (*read)(struct v_inode* inode, void* buffer, size_t len, size_t fpos);
+
+ // for operatiosn {write|read}_page, following are true:
+ // + `len` always equals to PG_SIZE
+ // + `fpos` always PG_SIZE aligned.
+ // These additional operations allow underlying fs to use more specialized
+ // and optimized code.
+
+ int (*write_page)(struct v_inode* inode, void* pg, size_t fpos);
+ int (*read_page)(struct v_inode* inode, void* pg, size_t fpos);
+
int (*readdir)(struct v_file* file, struct dir_context* dctx);
- int (*seek)(struct v_inode* inode, size_t offset); // optional
+ int (*seek)(struct v_file* file, size_t offset);
int (*close)(struct v_file* file);
int (*sync)(struct v_file* file);
};
struct v_inode_ops
{
- int (*create)(struct v_inode* this, struct v_dnode* dnode);
+ int (*create)(struct v_inode* this, struct v_dnode* dnode,
+ unsigned int itype);
+
int (*open)(struct v_inode* this, struct v_file* file);
int (*sync)(struct v_inode* this);
+
int (*mkdir)(struct v_inode* this, struct v_dnode* dnode);
- int (*rmdir)(struct v_inode* this, struct v_dnode* dir);
- int (*unlink)(struct v_inode* this);
+ int (*rmdir)(struct v_inode* this, struct v_dnode* dnode);
+ int (*unlink)(struct v_inode* this, struct v_dnode* name);
int (*link)(struct v_inode* this, struct v_dnode* new_name);
+
int (*read_symlink)(struct v_inode* this, const char** path_out);
int (*set_symlink)(struct v_inode* this, const char* target);
+
int (*dir_lookup)(struct v_inode* this, struct v_dnode* dnode);
+
int (*rename)(struct v_inode* from_inode,
struct v_dnode* from_dnode,
struct v_dnode* to_dnode);
+
int (*getxattr)(struct v_inode* this,
struct v_xattr_entry* entry); // optional
int (*setxattr)(struct v_inode* this,
struct v_inode* inode;
struct v_dnode* dnode;
struct llist_header* f_list;
- uint32_t f_pos;
- atomic_ulong ref_count;
+ u32_t f_pos;
+ unsigned long ref_count;
+ void* data;
struct v_file_ops* ops; // for caching
};
int flags;
};
+// [v_inode::aka_nodes]
+// how do we invalidate corresponding v_dnodes given the v_inode?
+/*
+ Consider taskfs, which is Lunaix's speak of Linux's procfs, that allow
+ info of every process being accessible via file system. Each process's
+ creation will result a creation of a directory under the root of task fs
+ with it's pid as name. But that dir must delete when process is killed, and
+ such deletion does not mediated by vfs itself, so there is a need of cache
+ syncing.
+ And this is also the case of all ramfs where argumentation to file tree is
+ performed by third party.
+*/
+
struct v_inode
{
inode_t id;
mutex_t lock;
- uint32_t itype;
+ u32_t itype;
time_t ctime;
time_t mtime;
time_t atime;
lba_t lb_addr;
- uint32_t open_count;
- uint32_t link_count;
- uint32_t lb_usage;
- uint32_t fsize;
+ u32_t open_count;
+ u32_t link_count;
+ u32_t lb_usage;
+ u32_t fsize;
+
+ u32_t acl;
+ uid_t uid;
+ gid_t gid;
+
+ void* data; // 允许底层FS绑定他的一些专有数据
+ struct llist_header aka_dnodes;
struct llist_header xattrs;
struct v_superblock* sb;
struct hlist_node hash_list;
struct lru_node lru;
struct pcache* pg_cache;
- void* data; // 允许底层FS绑定他的一些专有数据
struct v_inode_ops* ops;
struct v_file_ops* default_fops;
+
+ void (*destruct)(struct v_inode* inode);
};
struct v_mount
struct v_mount* parent;
struct v_dnode* mnt_point;
struct v_superblock* super_block;
- uint32_t busy_counter;
+ u32_t busy_counter;
int flags;
};
struct v_inode* inode;
struct v_dnode* parent;
struct hlist_node hash_list;
+ struct llist_header aka_list;
struct llist_header children;
struct llist_header siblings;
+
struct v_superblock* super_block;
struct v_mount* mnt;
- atomic_ulong ref_count;
+
+ unsigned long ref_count;
void* data;
+
+ void (*destruct)(struct v_dnode* dnode);
};
struct v_fdtable
struct pcache
{
struct v_inode* master;
- struct btrie tree;
- struct llist_header pages;
+ struct bcache cache;
struct llist_header dirty;
- uint32_t n_dirty;
- uint32_t n_pages;
+ u32_t n_dirty;
+ u32_t n_pages;
};
struct pcache_pg
{
- struct llist_header pg_list;
struct llist_header dirty_list;
- struct lru_node lru;
- struct pcache* holder;
- void* pg;
- uint32_t flags;
- uint32_t fpos;
+
+ union {
+ struct {
+ bool dirty:1;
+ };
+ u32_t flags;
+ };
+
+ void* data;
+ unsigned int index;
};
-/* --- file system manager --- */
+
+static inline bool
+check_itype_any(struct v_inode* inode, unsigned int type_mask)
+{
+ return !!(inode->itype & type_mask) || !type_mask;
+}
+
void
fsm_init();
struct filesystem*
fsm_get(const char* fs_name);
+void
+fsm_itbegin(struct fs_iter* iterator);
+
+bool
+fsm_itnext(struct fs_iter* iterator);
+
+static inline void
+fsm_itend(struct fs_iter* iterator)
+{
+ iterator->fs = NULL;
+}
+
void
vfs_init();
+void
+vfs_export_attributes();
+
struct v_dnode*
vfs_dcache_lookup(struct v_dnode* parent, struct hstr* str);
struct hstr* component,
int options);
+int
+vfs_walkat(int fd, const char* path, int at_opts,
+ struct v_dnode** dnode_out);
+
int
vfs_mount(const char* target,
const char* fs_name,
int
vfs_open(struct v_dnode* dnode, struct v_file** file);
+int
+vfs_pclose(struct v_file* file, pid_t pid);
+
int
vfs_close(struct v_file* file);
+void
+vfs_free_fd(struct v_fd* fd);
+
int
vfs_fsync(struct v_file* file);
vfs_sb_alloc();
void
-vfs_sb_free(struct v_superblock* sb);
+vfs_sb_unref(struct v_superblock* sb);
+
+void
+vfs_sb_ref(struct v_superblock* sb);
+
+#define vfs_assign_sb(sb_accessor, sb) \
+ ({ \
+ if (likely(sb_accessor)) { \
+ vfs_sb_unref(sb_accessor); \
+ } \
+ vfs_sb_ref(((sb_accessor) = (sb))); \
+ })
+
+static inline void
+vfs_i_assign_sb(struct v_inode* inode, struct v_superblock* sb)
+{
+ vfs_assign_sb(inode->sb, sb);
+}
+
+static inline void
+vfs_d_assign_sb(struct v_dnode* dnode, struct v_superblock* sb)
+{
+ vfs_assign_sb(dnode->super_block, sb);
+}
+
+static inline void
+vfs_d_assign_vmnt(struct v_dnode* dnode, struct v_mount* vmnt)
+{
+ if (dnode->mnt) {
+ assert_msg(dnode->mnt->mnt_point != dnode,
+ "vmnt must be detached first");
+ }
+
+ dnode->mnt = vmnt;
+
+ if (likely(vmnt))
+ vfs_d_assign_sb(dnode, vmnt->super_block);
+}
+
+static inline void
+vfs_vmnt_assign_sb(struct v_mount* vmnt, struct v_superblock* sb)
+{
+ vfs_assign_sb(vmnt->super_block, sb);
+}
struct v_dnode*
-vfs_d_alloc();
+vfs_d_alloc(struct v_dnode* parent, struct hstr* name);
void
vfs_d_free(struct v_dnode* dnode);
struct v_inode*
-vfs_i_find(struct v_superblock* sb, uint32_t i_id);
+vfs_i_find(struct v_superblock* sb, u32_t i_id);
void
vfs_i_addhash(struct v_inode* inode);
int
vfs_getfd(int fd, struct v_fd** fd_s);
-void
-pcache_init(struct pcache* pcache);
-
-void
-pcache_release_page(struct pcache* pcache, struct pcache_pg* page);
+int
+vfs_get_dtype(int itype);
-struct pcache_pg*
-pcache_new_page(struct pcache* pcache, uint32_t index);
+int
+vfs_get_path(struct v_dnode* dnode, char* buf, size_t size, int depth);
void
-pcache_set_dirty(struct pcache* pcache, struct pcache_pg* pg);
-
-struct pcache_pg*
-pcache_get_page(struct pcache* pcache,
- uint32_t index,
- uint32_t* offset,
- struct pcache_pg** page);
+pcache_init(struct pcache* pcache);
int
-pcache_write(struct v_inode* inode, void* data, uint32_t len, uint32_t fpos);
+pcache_write(struct v_inode* inode, void* data, u32_t len, u32_t fpos);
int
-pcache_read(struct v_inode* inode, void* data, uint32_t len, uint32_t fpos);
+pcache_read(struct v_inode* inode, void* data, u32_t len, u32_t fpos);
void
pcache_release(struct pcache* pcache);
int
vfs_mount_root(const char* fs_name, struct device* device);
-struct v_mount*
-vfs_create_mount(struct v_mount* parent, struct v_dnode* mnt_point);
+static inline bool
+mnt_check_busy(struct v_mount* mnt)
+{
+ return mnt->busy_counter > 1;
+}
+
+static inline void
+vfs_ref_dnode(struct v_dnode* dnode)
+{
+ dnode->ref_count++;
+
+ if (likely(dnode->mnt)) {
+ mnt_mkbusy(dnode->mnt);
+ }
+}
+
+static inline void
+vfs_unref_dnode(struct v_dnode* dnode)
+{
+ dnode->ref_count--;
+
+ if (likely(dnode->mnt)) {
+ mnt_chillax(dnode->mnt);
+ }
+}
+
+static inline void
+vfs_ref_file(struct v_file* file)
+{
+ file->ref_count++;
+}
+
+static inline void
+vfs_unref_file(struct v_file* file)
+{
+ file->ref_count--;
+}
+
+static inline bool
+vfs_check_duped_file(struct v_file* file)
+{
+ return file->ref_count > 1;
+}
int
vfs_check_writable(struct v_dnode* dnode);
size_t len,
size_t fpos);
+int
+default_file_read_page(struct v_inode* inode, void* buffer, size_t fpos);
+
+int
+default_file_write_page(struct v_inode* inode, void* buffer, size_t fpos);
+
int
default_file_readdir(struct v_file* file, struct dir_context* dctx);
default_file_close(struct v_file* file);
int
-default_file_seek(struct v_inode* inode, size_t offset);
+default_file_seek(struct v_file* file, size_t offset);
int
default_inode_open(struct v_inode* this, struct v_file* file);
void
xattr_addcache(struct v_inode* inode, struct v_xattr_entry* xattr);
+
+/* --- misc stuff --- */
+
+#define check_itype(to_check, itype) \
+ (((to_check) & (itype)) == (itype))
+
+/**
+ * @brief Check if node represent a regular file (nothing but a file)
+ *
+ * @param inode
+ * @return true
+ * @return false
+ */
+static inline bool
+check_regfile_node(struct v_inode* inode)
+{
+ return inode->itype == VFS_IFFILE;
+}
+
+/**
+ * @brief Check if node represent a file.
+ * This is basically everything within file system (dir, dev, etc.)
+ *
+ * @param inode
+ * @return true
+ * @return false
+ */
+static inline bool
+check_file_node(struct v_inode* inode)
+{
+ return check_itype(inode->itype, VFS_IFFILE);
+}
+
+static inline bool
+check_directory_node(struct v_inode* inode)
+{
+ return check_itype(inode->itype, VFS_IFDIR);
+}
+
+static inline bool
+check_device_node(struct v_inode* inode)
+{
+ return check_itype(inode->itype, VFS_IFDEV);
+}
+
+static inline bool
+check_seqdev_node(struct v_inode* inode)
+{
+ return check_device_node(inode);
+}
+
+static inline bool
+check_voldev_node(struct v_inode* inode)
+{
+ return check_itype(inode->itype, VFS_IFVOLDEV);
+}
+
+static inline bool
+check_symlink_node(struct v_inode* inode)
+{
+ return check_itype(inode->itype, VFS_IFSYMLINK);
+}
+
+static inline bool
+check_allow_ops(struct v_inode* inode, unsigned int perm)
+{
+ return fsacl_allow_ops(perm, inode->acl, inode->uid, inode->gid);
+}
+
+static inline bool
+check_allow_read(struct v_inode* inode)
+{
+ return check_allow_ops(inode, FSACL_aR);
+}
+
+static inline bool
+check_allow_write(struct v_inode* inode)
+{
+ return check_allow_ops(inode, FSACL_aW);
+}
+
+static inline bool
+check_allow_execute(struct v_inode* inode)
+{
+ return check_allow_ops(inode, FSACL_aX);
+}
+
#endif /* __LUNAIX_VFS_H */
git://scm.lunaixsky.com / lunaix-os.git / blobdiff