X-Git-Url: https://scm.lunaixsky.com/lunaix-os.git/blobdiff_plain/d57ee3ae693448387e3022fdd07bd741b2db818a..35a7d633d3f16c1e0539af6ca5d8e7482926cd93:/lunaix-os/kernel/fs/vfs.c diff --git a/lunaix-os/kernel/fs/vfs.c b/lunaix-os/kernel/fs/vfs.c index a69803e..0eec6bf 100644 --- a/lunaix-os/kernel/fs/vfs.c +++ b/lunaix-os/kernel/fs/vfs.c @@ -40,22 +40,23 @@ we can do similar thing in Lunaix? A block device emulation above the regular file when we mount it on. 10. (device) device number (dev_t) allocation - [good idea] :: composition + [good idea] :: composition (CHECKED) */ #include -#include #include #include #include -#include #include #include #include #include +#include #include +#include + static struct cake_pile* dnode_pile; static struct cake_pile* inode_pile; static struct cake_pile* file_pile; @@ -71,12 +72,6 @@ struct hstr vfs_ddot = HSTR("..", 2); struct hstr vfs_dot = HSTR(".", 1); struct hstr vfs_empty = HSTR("", 0); -struct v_superblock* -vfs_sb_alloc(); - -void -vfs_sb_free(struct v_superblock* sb); - static int __vfs_try_evict_dnode(struct lru_node* obj); @@ -96,29 +91,43 @@ vfs_init() dnode_cache = vzalloc(VFS_HASHTABLE_SIZE * sizeof(struct hbucket)); - dnode_lru = lru_new_zone(__vfs_try_evict_dnode); - inode_lru = lru_new_zone(__vfs_try_evict_inode); + dnode_lru = lru_new_zone("vfs_dnode", __vfs_try_evict_dnode); + inode_lru = lru_new_zone("vfs_inode", __vfs_try_evict_inode); hstr_rehash(&vfs_ddot, HSTR_FULL_HASH); hstr_rehash(&vfs_dot, HSTR_FULL_HASH); // 创建一个根dnode。 vfs_sysroot = vfs_d_alloc(NULL, &vfs_empty); + vfs_sysroot->parent = vfs_sysroot; atomic_fetch_add(&vfs_sysroot->ref_count, 1); } -inline struct hbucket* -__dcache_hash(struct v_dnode* parent, uint32_t* hash) +static inline struct hbucket* +__dcache_hash(struct v_dnode* parent, u32_t* hash) { - uint32_t _hash = *hash; - // 与parent的指针值做加法，来减小碰撞的可能性。 - _hash += (uint32_t)parent; + u32_t _hash = *hash; // 确保低位更加随机 _hash = _hash ^ (_hash >> VFS_HASHBITS); + // 与parent的指针值做加法，来减小碰撞的可能性。 + _hash += (u32_t)__ptr(parent); *hash = _hash; return &dnode_cache[_hash & VFS_HASH_MASK]; } +static inline int +__sync_inode_nolock(struct v_inode* inode) +{ + pcache_commit_all(inode); + + int errno = ENOTSUP; + if (inode->ops->sync) { + errno = inode->ops->sync(inode); + } + + return errno; +} + struct v_dnode* vfs_dcache_lookup(struct v_dnode* parent, struct hstr* str) { @@ -126,16 +135,16 @@ vfs_dcache_lookup(struct v_dnode* parent, struct hstr* str) return parent; if (HSTR_EQ(str, &vfs_ddot)) { - return parent->parent ? parent->parent : parent; + return parent->parent; } - uint32_t hash = str->hash; + u32_t hash = str->hash; struct hbucket* slot = __dcache_hash(parent, &hash); struct v_dnode *pos, *n; hashtable_bucket_foreach(slot, pos, n, hash_list) { - if (pos->name.hash == hash) { + if (pos->name.hash == hash && pos->parent == parent) { return pos; } } @@ -162,6 +171,7 @@ vfs_dcache_remove(struct v_dnode* dnode) assert(dnode->ref_count == 1); llist_delete(&dnode->siblings); + llist_delete(&dnode->aka_list); hlist_delete(&dnode->hash_list); dnode->parent = NULL; @@ -197,7 +207,7 @@ vfs_open(struct v_dnode* dnode, struct v_file** file) vfile->ref_count = ATOMIC_VAR_INIT(1); vfile->ops = inode->default_fops; - if ((inode->itype & VFS_IFFILE) && !inode->pg_cache) { + if (check_file_node(inode) && !inode->pg_cache) { struct pcache* pcache = vzalloc(sizeof(struct pcache)); pcache_init(pcache); pcache->master = inode; @@ -224,8 +234,11 @@ void vfs_assign_inode(struct v_dnode* assign_to, struct v_inode* inode) { if (assign_to->inode) { + llist_delete(&assign_to->aka_list); assign_to->inode->link_count--; } + + llist_append(&inode->aka_dnodes, &assign_to->aka_list); assign_to->inode = inode; inode->link_count++; } @@ -235,6 +248,10 @@ vfs_link(struct v_dnode* to_link, struct v_dnode* name) { int errno; + if ((errno = vfs_check_writable(to_link))) { + return errno; + } + lock_inode(to_link->inode); if (to_link->super_block->root != name->super_block->root) { errno = EXDEV; @@ -249,37 +266,105 @@ vfs_link(struct v_dnode* to_link, struct v_dnode* name) } int -vfs_close(struct v_file* file) +vfs_pclose(struct v_file* file, pid_t pid) { + struct v_inode* inode; int errno = 0; - if (!(errno = file->ops->close(file))) { - atomic_fetch_sub(&file->dnode->ref_count, 1); - file->inode->open_count--; - mnt_chillax(file->dnode->mnt); - pcache_commit_all(file->inode); - cake_release(file_pile, file); + inode = file->inode; + + /* + * Prevent dead lock. + * This happened when process is terminated while blocking on read. + * In that case, the process is still holding the inode lock and it + will never get released. + * The unlocking should also include ownership check. + * + * To see why, consider two process both open the same file both with + * fd=x. + * Process A: busy on reading x + * Process B: do nothing with x + * Assuming that, after a very short time, process B get terminated + * while process A is still busy in it's reading business. By this + * design, the inode lock of this file x is get released by B rather + * than A. And this will cause a probable race condition on A if other + * process is writing to this file later after B exit. + */ + + mutex_unlock_for(&inode->lock, pid); + + if (file->ref_count > 1) { + atomic_fetch_sub(&file->ref_count, 1); + return 0; + } + + if ((errno = file->ops->close(file))) { + goto done; + } + + atomic_fetch_sub(&file->dnode->ref_count, 1); + mnt_chillax(file->dnode->mnt); + cake_release(file_pile, file); + + /* + if the current inode is not being locked by other + threads that does not share same open context, + then we can try to do sync opportunistically + */ + if (mutex_on_hold(&inode->lock)) { + goto done; } + + lock_inode(inode); + + pcache_commit_all(inode); + inode->open_count--; + + if (!inode->open_count) { + __sync_inode_nolock(inode); + } + + unlock_inode(inode); + +done: return errno; } int -vfs_fsync(struct v_file* file) +vfs_close(struct v_file* file) { - lock_inode(file->inode); + return vfs_pclose(file, __current->pid); +} - int errno = ENOTSUP; - pcache_commit_all(file->inode); +void +vfs_free_fd(struct v_fd* fd) +{ + cake_release(fd_pile, fd); +} - if (file->ops->sync) { - errno = file->ops->sync(file); - } +int +vfs_isync(struct v_inode* inode) +{ + lock_inode(inode); - unlock_inode(file->inode); + int errno = __sync_inode_nolock(inode); + + unlock_inode(inode); return errno; } +int +vfs_fsync(struct v_file* file) +{ + int errno; + if ((errno = vfs_check_writable(file->dnode))) { + return errno; + } + + return vfs_isync(file->inode); +} + int vfs_alloc_fdslot(int* fd) { @@ -299,12 +384,30 @@ vfs_sb_alloc() memset(sb, 0, sizeof(*sb)); llist_init_head(&sb->sb_list); sb->i_cache = vzalloc(VFS_HASHTABLE_SIZE * sizeof(struct hbucket)); + sb->ref_count = 1; return sb; } +void +vfs_sb_ref(struct v_superblock* sb) +{ + sb->ref_count++; +} + void vfs_sb_free(struct v_superblock* sb) { + assert(sb->ref_count); + + sb->ref_count--; + if (sb->ref_count) { + return; + } + + if (sb->ops.release) { + sb->ops.release(sb); + } + vfree(sb->i_cache); cake_release(superblock_pile, sb); } @@ -348,6 +451,7 @@ vfs_d_alloc(struct v_dnode* parent, struct hstr* name) memset(dnode, 0, sizeof(*dnode)); llist_init_head(&dnode->children); llist_init_head(&dnode->siblings); + llist_init_head(&dnode->aka_list); mutex_init(&dnode->lock); dnode->ref_count = ATOMIC_VAR_INIT(0); @@ -356,7 +460,8 @@ vfs_d_alloc(struct v_dnode* parent, struct hstr* name) hstrcpy(&dnode->name, name); if (parent) { - dnode->super_block = parent->super_block; + vfs_d_assign_sb(dnode, parent->super_block); + dnode->mnt = parent->mnt; } lru_use_one(dnode_lru, &dnode->lru); @@ -384,12 +489,17 @@ vfs_d_free(struct v_dnode* dnode) vfs_dcache_remove(pos); } - vfree(dnode->name.value); + if (dnode->destruct) { + dnode->destruct(dnode); + } + + vfs_sb_free(dnode->super_block); + vfree((void*)dnode->name.value); cake_release(dnode_pile, 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) { struct hbucket* slot = &sb->i_cache[i_id & VFS_HASH_MASK]; struct v_inode *pos, *n; @@ -429,15 +539,15 @@ vfs_i_alloc(struct v_superblock* sb) memset(inode, 0, sizeof(*inode)); mutex_init(&inode->lock); llist_init_head(&inode->xattrs); + llist_init_head(&inode->aka_dnodes); sb->ops.init_inode(sb, inode); - inode->sb = sb; inode->ctime = clock_unixtime(); inode->atime = inode->ctime; inode->mtime = inode->ctime; -done: + vfs_i_assign_sb(inode, sb); lru_use_one(inode_lru, &inode->lru); return inode; } @@ -449,14 +559,28 @@ vfs_i_free(struct v_inode* inode) pcache_release(inode->pg_cache); vfree(inode->pg_cache); } - inode->ops->sync(inode); + // we don't need to sync inode. + // If an inode can be free, then it must be properly closed. + // Hence it must be synced already! + if (inode->destruct) { + inode->destruct(inode); + } + + vfs_sb_free(inode->sb); hlist_delete(&inode->hash_list); cake_release(inode_pile, inode); } /* ---- System call definition and support ---- */ -#define FLOCATE_CREATE_EMPTY 1 +// make a new name when not exists +#define FLOC_MAYBE_MKNAME 1 + +// name must be non-exist and made. +#define FLOC_MKNAME 2 + +// no follow symlink +#define FLOC_NOFOLLOW 4 int vfs_getfd(int fd, struct v_fd** fd_s) @@ -467,42 +591,104 @@ vfs_getfd(int fd, struct v_fd** fd_s) return EBADF; } -int +static int +__vfs_mknod(struct v_inode* parent, struct v_dnode* dnode, + unsigned int itype, dev_t* dev) +{ + int errno; + + errno = parent->ops->create(parent, dnode, itype); + if (errno) { + return errno; + } + + return 0; +} + +struct file_locator { + struct v_dnode* dir; + struct v_dnode* file; + bool fresh; +}; + +/** + * @brief unlock the file locator (floc) if possible. + * If the file to be located if not exists, and + * any FLOC_*MKNAME flag is set, then the parent + * dnode will be locked until the file has been properly + * finalised by subsequent logic. + * + * @param floc + */ +static inline void +__floc_try_unlock(struct file_locator* floc) +{ + if (floc->fresh) { + assert(floc->dir); + unlock_dnode(floc->dir); + } +} + +static int __vfs_try_locate_file(const char* path, - struct v_dnode** fdir, - struct v_dnode** file, + struct file_locator* floc, int options) { char name_str[VFS_NAME_MAXLEN]; + struct v_dnode *fdir, *file; struct hstr name = HSTR(name_str, 0); - int errno; - if ((errno = - vfs_walk(__current->cwd, path, fdir, &name, VFS_WALK_PARENT))) { - return errno; + int errno, woption = 0; + + if ((options & FLOC_NOFOLLOW)) { + woption |= VFS_WALK_NOFOLLOW; + options &= ~FLOC_NOFOLLOW; } - errno = vfs_walk(*fdir, name.value, file, NULL, 0); - if (errno != ENOENT || !(options & FLOCATE_CREATE_EMPTY)) { + floc->fresh = false; + name_str[0] = 0; + errno = vfs_walk_proc(path, &fdir, &name, woption | VFS_WALK_PARENT); + if (errno) { return errno; } - struct v_dnode* parent = *fdir; - struct v_dnode* file_new = vfs_d_alloc(parent, &name); + errno = vfs_walk(fdir, name.value, &file, NULL, woption); - if (!file_new) { - return ENOMEM; + if (errno && errno != ENOENT) { + goto done; + } + + if (!errno) { + if ((options & FLOC_MKNAME)) { + errno = EEXIST; + } + goto done; } - lock_dnode(parent); + // errno == ENOENT + if (!options) { + goto done; + } - if (!(errno = parent->inode->ops->create(parent->inode, file_new))) { - vfs_dcache_add(parent, file_new); - *file = file_new; - } else { - vfs_d_free(file_new); + errno = vfs_check_writable(fdir); + if (errno) { + goto done; } - unlock_dnode(parent); + floc->fresh = true; + + file = vfs_d_alloc(fdir, &name); + + if (!file) { + return ENOMEM; + } + + lock_dnode(fdir); + + vfs_dcache_add(fdir, file); + +done: + floc->dir = fdir; + floc->file = file; return errno; } @@ -510,29 +696,64 @@ __vfs_try_locate_file(const char* path, int vfs_do_open(const char* path, int options) { - int errno, fd; + int errno, fd, loptions = 0; struct v_dnode *dentry, *file; - struct v_file* ofile = 0; + struct v_file* ofile = NULL; + struct file_locator floc; + struct v_inode* inode; + + if ((options & FO_CREATE)) { + loptions |= FLOC_MAYBE_MKNAME; + } else if ((options & FO_NOFOLLOW)) { + loptions |= FLOC_NOFOLLOW; + } - errno = __vfs_try_locate_file( - path, &dentry, &file, (options & FO_CREATE) ? FLOCATE_CREATE_EMPTY : 0); + errno = __vfs_try_locate_file(path, &floc, loptions); - if (errno || (errno = vfs_open(file, &ofile))) { + if (errno || (errno = vfs_alloc_fdslot(&fd))) { return errno; } - struct v_inode* o_inode = ofile->inode; + file = floc.file; + dentry = floc.dir; + + if (floc.fresh) { + errno = __vfs_mknod(dentry->inode, file, VFS_IFFILE, NULL); + if (errno) { + vfs_d_free(file); + __floc_try_unlock(&floc); + return errno; + } - if (!errno && !(errno = vfs_alloc_fdslot(&fd))) { - struct v_fd* fd_s = vzalloc(sizeof(*fd_s)); - ofile->f_pos = ofile->inode->fsize & -((options & FO_APPEND) != 0); - fd_s->file = ofile; - fd_s->flags = options; - __current->fdtable->fds[fd] = fd_s; - return fd; + __floc_try_unlock(&floc); } - return errno; + + if ((errno = vfs_open(file, &ofile))) { + return errno; + } + + inode = ofile->inode; + lock_inode(inode); + + struct v_fd* fd_s = cake_grab(fd_pile); + memset(fd_s, 0, sizeof(*fd_s)); + + if ((options & O_TRUNC)) { + file->inode->fsize = 0; + } + + if (vfs_get_dtype(inode->itype) == DT_DIR) { + ofile->f_pos = 0; + } + + fd_s->file = ofile; + fd_s->flags = options; + __current->fdtable->fds[fd] = fd_s; + + unlock_inode(inode); + + return fd; } __DEFINE_LXSYSCALL2(int, open, const char*, path, int, options) @@ -549,13 +770,11 @@ __DEFINE_LXSYSCALL1(int, close, int, fd) goto done_err; } - if (fd_s->file->ref_count > 1) { - fd_s->file->ref_count--; - } else if ((errno = vfs_close(fd_s->file))) { + if ((errno = vfs_close(fd_s->file))) { goto done_err; } - vfree(fd_s); + cake_release(fd_pile, fd_s); __current->fdtable->fds[fd] = 0; done_err: @@ -568,13 +787,13 @@ __vfs_readdir_callback(struct dir_context* dctx, const int len, const int dtype) { - struct dirent* dent = (struct dirent*)dctx->cb_data; - strncpy(dent->d_name, name, DIRENT_NAME_MAX_LEN); + struct lx_dirent* dent = (struct lx_dirent*)dctx->cb_data; + strncpy(dent->d_name, name, MIN(len, DIRENT_NAME_MAX_LEN)); dent->d_nlen = len; dent->d_type = dtype; } -__DEFINE_LXSYSCALL2(int, readdir, int, fd, struct dirent*, dent) +__DEFINE_LXSYSCALL2(int, sys_readdir, int, fd, struct lx_dirent*, dent) { struct v_fd* fd_s; int errno; @@ -587,29 +806,23 @@ __DEFINE_LXSYSCALL2(int, readdir, int, fd, struct dirent*, dent) lock_inode(inode); - if (!(inode->itype & VFS_IFDIR)) { + if (!check_directory_node(inode)) { errno = ENOTDIR; - } else { - struct dir_context dctx = - (struct dir_context){ .cb_data = dent, - .index = dent->d_offset, - .read_complete_callback = - __vfs_readdir_callback }; - errno = 1; - if (dent->d_offset == 0) { - __vfs_readdir_callback(&dctx, vfs_dot.value, vfs_dot.len, 0); - } else if (dent->d_offset == 1) { - __vfs_readdir_callback(&dctx, vfs_ddot.value, vfs_ddot.len, 0); - } else { - dctx.index -= 2; - if ((errno = fd_s->file->ops->readdir(fd_s->file, &dctx)) != 1) { - unlock_inode(inode); - goto done; - } - } - dent->d_offset++; + goto unlock; } + struct dir_context dctx = (struct dir_context) { + .cb_data = dent, + .read_complete_callback = __vfs_readdir_callback + }; + + if ((errno = fd_s->file->ops->readdir(fd_s->file, &dctx)) != 1) { + goto unlock; + } + dent->d_offset++; + fd_s->file->f_pos++; + +unlock: unlock_inode(inode); done: @@ -625,7 +838,7 @@ __DEFINE_LXSYSCALL3(int, read, int, fd, void*, buf, size_t, count) } struct v_file* file = fd_s->file; - if ((file->inode->itype & VFS_IFDIR)) { + if (check_directory_node(file->inode)) { errno = EISDIR; goto done; } @@ -634,13 +847,11 @@ __DEFINE_LXSYSCALL3(int, read, int, fd, void*, buf, size_t, count) file->inode->atime = clock_unixtime(); - __SYSCALL_INTERRUPTIBLE({ - if ((file->inode->itype & VFS_IFSEQDEV) || (fd_s->flags & FO_DIRECT)) { - errno = file->ops->read(file->inode, buf, count, file->f_pos); - } else { - errno = pcache_read(file->inode, buf, count, file->f_pos); - } - }) + if (check_seqdev_node(file->inode) || (fd_s->flags & FO_DIRECT)) { + errno = file->ops->read(file->inode, buf, count, file->f_pos); + } else { + errno = pcache_read(file->inode, buf, count, file->f_pos); + } if (errno > 0) { file->f_pos += errno; @@ -662,31 +873,41 @@ __DEFINE_LXSYSCALL3(int, write, int, fd, void*, buf, size_t, count) goto done; } + struct v_inode* inode; struct v_file* file = fd_s->file; - if ((file->inode->itype & VFS_IFDIR)) { + + if ((errno = vfs_check_writable(file->dnode))) { + goto done; + } + + if (check_directory_node(file->inode)) { errno = EISDIR; goto done; } - lock_inode(file->inode); + inode = file->inode; + lock_inode(inode); - file->inode->mtime = clock_unixtime(); + inode->mtime = clock_unixtime(); + if ((fd_s->flags & O_APPEND)) { + file->f_pos = inode->fsize; + } - __SYSCALL_INTERRUPTIBLE({ - if ((file->inode->itype & VFS_IFSEQDEV) || (fd_s->flags & FO_DIRECT)) { - errno = file->ops->write(file->inode, buf, count, file->f_pos); - } else { - errno = pcache_write(file->inode, buf, count, file->f_pos); - } - }) + if (check_seqdev_node(inode) || (fd_s->flags & FO_DIRECT)) { + errno = file->ops->write(inode, buf, count, file->f_pos); + } else { + errno = pcache_write(inode, buf, count, file->f_pos); + } if (errno > 0) { file->f_pos += errno; - unlock_inode(file->inode); + inode->fsize = MAX(inode->fsize, file->f_pos); + + unlock_inode(inode); return errno; } - unlock_inode(file->inode); + unlock_inode(inode); done: return DO_STATUS(errno); @@ -701,26 +922,42 @@ __DEFINE_LXSYSCALL3(int, lseek, int, fd, int, offset, int, options) } struct v_file* file = fd_s->file; + struct v_inode* inode = file->inode; - lock_inode(file->inode); + if (!file->ops->seek) { + errno = ENOTSUP; + goto done; + } + + lock_inode(inode); + + int overflow = 0; + int fpos = file->f_pos; - size_t fpos = file->f_pos; + if (vfs_get_dtype(inode->itype) == DT_DIR) { + options = (options != FSEEK_END) ? options : FSEEK_SET; + } + switch (options) { case FSEEK_CUR: - fpos = (size_t)((int)file->f_pos + offset); + overflow = sadd_of((int)file->f_pos, offset, &fpos); break; case FSEEK_END: - fpos = (size_t)((int)file->inode->fsize + offset); + overflow = sadd_of((int)inode->fsize, offset, &fpos); break; case FSEEK_SET: fpos = offset; break; } - if (!(errno = file->ops->seek(file->inode, fpos))) { - file->f_pos = fpos; + + if (overflow) { + errno = EOVERFLOW; + } + else { + errno = file->ops->seek(file, fpos); } - unlock_inode(file->inode); + unlock_inode(inode); done: return DO_STATUS(errno); @@ -734,23 +971,27 @@ vfs_get_path(struct v_dnode* dnode, char* buf, size_t size, int depth) } if (depth > 64) { - return ELOOP; + return ENAMETOOLONG; } - size_t len = vfs_get_path(dnode->parent, buf, size, depth + 1); + size_t len = 0; + + if (dnode->parent != dnode) { + len = vfs_get_path(dnode->parent, buf, size, depth + 1); + } if (len >= size) { return len; } + if (!len || buf[len - 1] != VFS_PATH_DELIM) { + buf[len++] = VFS_PATH_DELIM; + } + size_t cpy_size = MIN(dnode->name.len, size - len); strncpy(buf + len, dnode->name.value, cpy_size); len += cpy_size; - if (len < size) { - buf[len++] = VFS_PATH_DELIM; - } - return len; } @@ -759,16 +1000,42 @@ vfs_readlink(struct v_dnode* dnode, char* buf, size_t size) { const char* link; struct v_inode* inode = dnode->inode; - if (inode->ops->read_symlink) { - lock_inode(inode); - int errno = inode->ops->read_symlink(inode, &link); - strncpy(buf, link, size); + if (!check_symlink_node(inode)) { + return EINVAL; + } - unlock_inode(inode); - return errno; + if (!inode->ops->read_symlink) { + return ENOTSUP; } - return 0; + + lock_inode(inode); + + int errno = inode->ops->read_symlink(inode, &link); + if (errno >= 0) { + strncpy(buf, link, MIN(size, (size_t)errno)); + } + + unlock_inode(inode); + return errno; +} + +int +vfs_get_dtype(int itype) +{ + int dtype = DT_FILE; + if (check_itype(itype, VFS_IFSYMLINK)) { + dtype |= DT_SYMLINK; + } + + if (check_itype(itype, VFS_IFDIR)) { + dtype |= DT_DIR; + return dtype; + } + + // TODO other types + + return dtype; } __DEFINE_LXSYSCALL3(int, realpathat, int, fd, char*, buf, size_t, size) @@ -794,8 +1061,7 @@ __DEFINE_LXSYSCALL3(int, readlink, const char*, path, char*, buf, size_t, size) { int errno; struct v_dnode* dnode; - if (!(errno = - vfs_walk(__current->cwd, path, &dnode, NULL, VFS_WALK_NOFOLLOW))) { + if (!(errno = vfs_walk_proc(path, &dnode, NULL, VFS_WALK_NOFOLLOW))) { errno = vfs_readlink(dnode, buf, size); } @@ -806,16 +1072,8 @@ __DEFINE_LXSYSCALL3(int, readlink, const char*, path, char*, buf, size_t, size) return DO_STATUS(errno); } -__DEFINE_LXSYSCALL4(int, - readlinkat, - int, - dirfd, - const char*, - pathname, - char*, - buf, - size_t, - size) +__DEFINE_LXSYSCALL4( + int, readlinkat, int, dirfd, const char*, pathname, char*, buf, size_t, size) { int errno; struct v_fd* fd_s; @@ -823,6 +1081,8 @@ __DEFINE_LXSYSCALL4(int, goto done; } + pathname = pathname ? pathname : ""; + struct v_dnode* dnode; if (!(errno = vfs_walk( fd_s->file->dnode, pathname, &dnode, NULL, VFS_WALK_NOFOLLOW))) { @@ -849,12 +1109,16 @@ __DEFINE_LXSYSCALL1(int, rmdir, const char*, pathname) { int errno; struct v_dnode* dnode; - if ((errno = vfs_walk(__current->cwd, pathname, &dnode, NULL, 0))) { + if ((errno = vfs_walk_proc(pathname, &dnode, NULL, 0))) { return DO_STATUS(errno); } lock_dnode(dnode); + if ((errno = vfs_check_writable(dnode))) { + goto done; + } + if ((dnode->super_block->fs->types & FSTYPE_ROFS)) { errno = EROFS; goto done; @@ -880,7 +1144,7 @@ __DEFINE_LXSYSCALL1(int, rmdir, const char*, pathname) lock_dnode(parent); lock_inode(parent->inode); - if ((dnode->inode->itype & VFS_IFDIR)) { + if (check_directory_node(dnode->inode)) { errno = parent->inode->ops->rmdir(parent->inode, dnode); if (!errno) { vfs_dcache_remove(dnode); @@ -904,28 +1168,36 @@ __DEFINE_LXSYSCALL1(int, mkdir, const char*, path) char name_value[VFS_NAME_MAXLEN]; struct hstr name = HHSTR(name_value, 0, 0); - if (!dir) { - errno = ENOMEM; + if ((errno = vfs_walk_proc(path, &parent, &name, VFS_WALK_PARENT))) { goto done; } - if ((errno = - vfs_walk(__current->cwd, path, &parent, &name, VFS_WALK_PARENT))) { + if (!(errno = vfs_walk(parent, name_value, &dir, NULL, 0))) { + errno = EEXIST; + goto done; + } + + if ((errno = vfs_check_writable(parent))) { goto done; } - dir = vfs_d_alloc(parent, &name); + if (!(dir = vfs_d_alloc(parent, &name))) { + errno = ENOMEM; + goto done; + } + + struct v_inode* inode = parent->inode; lock_dnode(parent); - lock_inode(parent->inode); + lock_inode(inode); if ((parent->super_block->fs->types & FSTYPE_ROFS)) { errno = ENOTSUP; - } else if (!parent->inode->ops->mkdir) { + } else if (!inode->ops->mkdir) { errno = ENOTSUP; - } else if (!(parent->inode->itype & VFS_IFDIR)) { + } else if (!check_directory_node(inode)) { errno = ENOTDIR; - } else if (!(errno = parent->inode->ops->mkdir(parent->inode, dir))) { + } else if (!(errno = inode->ops->mkdir(inode, dir))) { vfs_dcache_add(parent, dir); goto cleanup; } @@ -933,7 +1205,7 @@ __DEFINE_LXSYSCALL1(int, mkdir, const char*, path) vfs_d_free(dir); cleanup: - unlock_inode(parent->inode); + unlock_inode(inode); unlock_dnode(parent); done: return DO_STATUS(errno); @@ -942,21 +1214,23 @@ done: int __vfs_do_unlink(struct v_dnode* dnode) { + int errno; struct v_inode* inode = dnode->inode; if (dnode->ref_count > 1) { return EBUSY; } + if ((errno = vfs_check_writable(dnode))) { + return errno; + } + lock_inode(inode); - int errno; if (inode->open_count) { errno = EBUSY; - } else if (!(inode->itype & VFS_IFDIR)) { - // The underlying unlink implementation should handle - // symlink case - errno = inode->ops->unlink(inode); + } else if (!check_directory_node(inode)) { + errno = inode->ops->unlink(inode, dnode); if (!errno) { vfs_d_free(dnode); } @@ -973,11 +1247,7 @@ __DEFINE_LXSYSCALL1(int, unlink, const char*, pathname) { int errno; struct v_dnode* dnode; - if ((errno = vfs_walk(__current->cwd, pathname, &dnode, NULL, 0))) { - goto done; - } - if ((dnode->super_block->fs->types & FSTYPE_ROFS)) { - errno = EROFS; + if ((errno = vfs_walk_proc(pathname, &dnode, NULL, 0))) { goto done; } @@ -1007,18 +1277,30 @@ done: __DEFINE_LXSYSCALL2(int, link, const char*, oldpath, const char*, newpath) { int errno; - struct v_dnode *dentry, *to_link, *name_dentry, *name_file; + struct file_locator floc; + struct v_dnode *to_link, *name_file; + + errno = __vfs_try_locate_file(oldpath, &floc, 0); + if (errno) { + goto done; + } + + __floc_try_unlock(&floc); - errno = __vfs_try_locate_file(oldpath, &dentry, &to_link, 0); + to_link = floc.file; + errno = __vfs_try_locate_file(newpath, &floc, FLOC_MKNAME); if (!errno) { - errno = __vfs_try_locate_file( - newpath, &name_dentry, &name_file, FLOCATE_CREATE_EMPTY); - if (!errno) { - errno = EEXIST; - } else if (name_file) { - errno = vfs_link(to_link, name_file); - } + goto done; } + + name_file = floc.file; + errno = vfs_link(to_link, name_file); + if (errno) { + vfs_d_free(name_file); + } + +done: + __floc_try_unlock(&floc); return DO_STATUS(errno); } @@ -1026,6 +1308,7 @@ __DEFINE_LXSYSCALL1(int, fsync, int, fildes) { int errno; struct v_fd* fd_s; + if (!(errno = vfs_getfd(fildes, &fd_s))) { errno = vfs_fsync(fd_s->file); } @@ -1103,57 +1386,94 @@ done: return DO_STATUS(errno); } -__DEFINE_LXSYSCALL2(int, - symlink, - const char*, - pathname, - const char*, - link_target) +__DEFINE_LXSYSCALL2( + int, symlink, const char*, pathname, const char*, link_target) { int errno; - struct v_dnode* dnode; - if ((errno = vfs_walk(__current->cwd, pathname, &dnode, NULL, 0))) { + struct file_locator floc; + struct v_dnode *file; + struct v_inode *f_ino; + + errno = __vfs_try_locate_file(pathname, &floc, FLOC_MKNAME); + if (errno) { goto done; } - if ((dnode->super_block->fs->types & FSTYPE_ROFS)) { - errno = EROFS; + + file = floc.file; + errno = __vfs_mknod(floc.dir->inode, file, VFS_IFSYMLINK, NULL); + if (errno) { + vfs_d_free(file); goto done; } - if (!dnode->inode->ops->set_symlink) { + + f_ino = file->inode; + + assert(f_ino); + + errno = vfs_check_writable(file); + if (errno) { + goto done; + } + + if (!f_ino->ops->set_symlink) { errno = ENOTSUP; goto done; } - lock_inode(dnode->inode); + lock_inode(f_ino); - errno = dnode->inode->ops->set_symlink(dnode->inode, link_target); + errno = f_ino->ops->set_symlink(f_ino, link_target); - unlock_inode(dnode->inode); + unlock_inode(f_ino); done: + __floc_try_unlock(&floc); return DO_STATUS(errno); } +void +vfs_ref_file(struct v_file* file) +{ + atomic_fetch_add(&file->ref_count, 1); +} + +void +vfs_ref_dnode(struct v_dnode* dnode) +{ + atomic_fetch_add(&dnode->ref_count, 1); + + if (dnode->mnt) { + mnt_mkbusy(dnode->mnt); + } +} + +void +vfs_unref_dnode(struct v_dnode* dnode) +{ + atomic_fetch_sub(&dnode->ref_count, 1); + if (dnode->mnt) { + mnt_chillax(dnode->mnt); + } +} + int -__vfs_do_chdir(struct v_dnode* dnode) +vfs_do_chdir(struct proc_info* proc, struct v_dnode* dnode) { int errno = 0; lock_dnode(dnode); - if (!(dnode->inode->itype & VFS_IFDIR)) { + if (!check_directory_node(dnode->inode)) { errno = ENOTDIR; goto done; } - if (__current->cwd) { - atomic_fetch_sub(&__current->cwd->ref_count, 1); - mnt_chillax(__current->cwd->mnt); + if (proc->cwd) { + vfs_unref_dnode(proc->cwd); } - atomic_fetch_add(&dnode->ref_count, 1); - mnt_mkbusy(dnode->mnt); - __current->cwd = dnode; + vfs_ref_dnode(dnode); + proc->cwd = dnode; unlock_dnode(dnode); @@ -1166,11 +1486,11 @@ __DEFINE_LXSYSCALL1(int, chdir, const char*, path) struct v_dnode* dnode; int errno = 0; - if ((errno = vfs_walk(__current->cwd, path, &dnode, NULL, 0))) { + if ((errno = vfs_walk_proc(path, &dnode, NULL, 0))) { goto done; } - errno = __vfs_do_chdir(dnode); + errno = vfs_do_chdir((struct proc_info*)__current, dnode); done: return DO_STATUS(errno); @@ -1185,7 +1505,7 @@ __DEFINE_LXSYSCALL1(int, fchdir, int, fd) goto done; } - errno = __vfs_do_chdir(fd_s->file->dnode); + errno = vfs_do_chdir((struct proc_info*)__current, fd_s->file->dnode); done: return DO_STATUS(errno); @@ -1213,23 +1533,28 @@ __DEFINE_LXSYSCALL2(char*, getcwd, char*, buf, size_t, size) } } - buf[len + 1] = '\0'; + buf[len] = '\0'; ret_ptr = buf; done: - __current->k_status = errno; + syscall_result(errno); return ret_ptr; } int vfs_do_rename(struct v_dnode* current, struct v_dnode* target) { + int errno = 0; if (current->inode->id == target->inode->id) { // hard link return 0; } + if ((errno = vfs_check_writable(current))) { + return errno; + } + if (current->ref_count > 1 || target->ref_count > 1) { return EBUSY; } @@ -1238,8 +1563,6 @@ vfs_do_rename(struct v_dnode* current, struct v_dnode* target) return EXDEV; } - int errno = 0; - struct v_dnode* oldparent = current->parent; struct v_dnode* newparent = target->parent; @@ -1287,7 +1610,7 @@ __DEFINE_LXSYSCALL2(int, rename, const char*, oldpath, const char*, newpath) struct hstr name = HSTR(valloc(VFS_NAME_MAXLEN), 0); int errno = 0; - if ((errno = vfs_walk(__current->cwd, oldpath, &cur, NULL, 0))) { + if ((errno = vfs_walk_proc(oldpath, &cur, NULL, 0))) { goto done; } @@ -1312,6 +1635,47 @@ __DEFINE_LXSYSCALL2(int, rename, const char*, oldpath, const char*, newpath) errno = vfs_do_rename(cur, target); done: - vfree(name.value); + vfree((void*)name.value); + return DO_STATUS(errno); +} + +__DEFINE_LXSYSCALL2(int, fstat, int, fd, struct file_stat*, stat) +{ + int errno = 0; + struct v_fd* fds; + + if ((errno = vfs_getfd(fd, &fds))) { + goto done; + } + + struct v_inode* vino = fds->file->inode; + struct device* fdev = vino->sb->dev; + + *stat = (struct file_stat){.st_ino = vino->id, + .st_blocks = vino->lb_usage, + .st_size = vino->fsize, + .mode = vino->itype, + .st_ioblksize = PAGE_SIZE, + .st_blksize = vino->sb->blksize}; + + if (check_device_node(vino)) { + struct device* rdev = resolve_device(vino->data); + if (!rdev) { + errno = EINVAL; + goto done; + } + + stat->st_rdev = (dev_t){.meta = rdev->ident.fn_grp, + .unique = rdev->ident.unique, + .index = dev_uid(rdev) }; + } + + if (fdev) { + stat->st_dev = (dev_t){.meta = fdev->ident.fn_grp, + .unique = fdev->ident.unique, + .index = dev_uid(fdev) }; + } + +done: return DO_STATUS(errno); } \ No newline at end of file