Boot framework rework (#45)

[lunaix-os.git] / lunaix-os / kernel / fs / vfs.c

diff --git a/lunaix-os/kernel/fs/vfs.c b/lunaix-os/kernel/fs/vfs.c
index 3592dc8e2fd9a43e43637b131d0929fca1fb4782..025fa3ff438ef8febaa8a8e48f9096bdf075fc4f 100644 (file)
--- 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
                     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] <class>:<subclass>:<uniq_id> composition
+            [good idea] <class>:<subclass>:<uniq_id> composition (CHECKED)

 */
 
 #include <klibc/string.h>
 */
 
 #include <klibc/string.h>

-#include <lunaix/dirent.h>

 #include <lunaix/foptions.h>
 #include <lunaix/fs.h>
 #include <lunaix/mm/cake.h>
 #include <lunaix/foptions.h>
 #include <lunaix/fs.h>
 #include <lunaix/mm/cake.h>

-#include <lunaix/mm/page.h>

 #include <lunaix/mm/valloc.h>
 #include <lunaix/process.h>
 #include <lunaix/spike.h>
 #include <lunaix/syscall.h>
 #include <lunaix/mm/valloc.h>
 #include <lunaix/process.h>
 #include <lunaix/spike.h>
 #include <lunaix/syscall.h>

+#include <lunaix/syscall_utils.h>

 
 #include <lunaix/fs/twifs.h>
 
 
 #include <lunaix/fs/twifs.h>
 

+#include <usr/lunaix/dirent_defs.h>
+

 static struct cake_pile* dnode_pile;
 static struct cake_pile* inode_pile;
 static struct cake_pile* file_pile;
 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 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);
 
 static int
 __vfs_try_evict_dnode(struct lru_node* obj);
 


@@ -96,8 +91,8 @@ vfs_init()
 
     dnode_cache = vzalloc(VFS_HASHTABLE_SIZE * sizeof(struct hbucket));
 
 
     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);
 
     hstr_rehash(&vfs_ddot, HSTR_FULL_HASH);
     hstr_rehash(&vfs_dot, HSTR_FULL_HASH);


@@ -108,18 +103,31 @@ vfs_init()
     atomic_fetch_add(&vfs_sysroot->ref_count, 1);
 }
 
     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);
     // 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];
 }
 
     *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)
 {
 struct v_dnode*
 vfs_dcache_lookup(struct v_dnode* parent, struct hstr* str)
 {


@@ -130,13 +138,13 @@ vfs_dcache_lookup(struct v_dnode* parent, struct hstr* str)
         return 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)
     {
     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;
         }
     }
             return pos;
         }
     }


@@ -163,6 +171,7 @@ vfs_dcache_remove(struct v_dnode* dnode)
     assert(dnode->ref_count == 1);
 
     llist_delete(&dnode->siblings);
     assert(dnode->ref_count == 1);
 
     llist_delete(&dnode->siblings);

+    llist_delete(&dnode->aka_list);

     hlist_delete(&dnode->hash_list);
 
     dnode->parent = NULL;
     hlist_delete(&dnode->hash_list);
 
     dnode->parent = NULL;


@@ -198,7 +207,7 @@ vfs_open(struct v_dnode* dnode, struct v_file** file)
     vfile->ref_count = ATOMIC_VAR_INIT(1);
     vfile->ops = inode->default_fops;
 
     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;
         struct pcache* pcache = vzalloc(sizeof(struct pcache));
         pcache_init(pcache);
         pcache->master = inode;


@@ -225,8 +234,11 @@ void
 vfs_assign_inode(struct v_dnode* assign_to, struct v_inode* inode)
 {
     if (assign_to->inode) {
 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--;
     }
         assign_to->inode->link_count--;
     }

+
+    llist_append(&inode->aka_dnodes, &assign_to->aka_list);

     assign_to->inode = inode;
     inode->link_count++;
 }
     assign_to->inode = inode;
     inode->link_count++;
 }


@@ -254,42 +266,105 @@ vfs_link(struct v_dnode* to_link, struct v_dnode* name)
 }
 
 int
 }
 
 int

-vfs_close(struct v_file* file)
+vfs_pclose(struct v_file* file, pid_t pid)

 {
 {

+    struct v_inode* inode;

     int errno = 0;
     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
     return errno;
 }
 
 int

-vfs_fsync(struct v_file* file)
+vfs_close(struct v_file* file)

 {
 {

-    int errno;
-    if ((errno = vfs_check_writable(file->dnode))) {
-        return errno;
-    }
+    return vfs_pclose(file, __current->pid);
+}

 
 

-    lock_inode(file->inode);
+void
+vfs_free_fd(struct v_fd* fd)
+{
+    cake_release(fd_pile, fd);
+}

 
 

-    pcache_commit_all(file->inode);
+int
+vfs_isync(struct v_inode* inode)
+{
+    lock_inode(inode);

 
 

-    errno = ENOTSUP;
-    if (file->ops->sync) {
-        errno = file->ops->sync(file);
-    }
+    int errno = __sync_inode_nolock(inode);

 
 

-    unlock_inode(file->inode);
+    unlock_inode(inode);

 
     return errno;
 }
 
 
     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)
 {
 int
 vfs_alloc_fdslot(int* fd)
 {


@@ -309,12 +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));
     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;
 }
 
     return sb;
 }
 

+void
+vfs_sb_ref(struct v_superblock* sb)
+{
+    sb->ref_count++;
+}
+

 void
 vfs_sb_free(struct v_superblock* sb)
 {
 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);
 }
     vfree(sb->i_cache);
     cake_release(superblock_pile, sb);
 }


@@ -358,6 +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);
     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);
     mutex_init(&dnode->lock);
 
     dnode->ref_count = ATOMIC_VAR_INIT(0);


@@ -366,7 +460,7 @@ vfs_d_alloc(struct v_dnode* parent, struct hstr* name)
     hstrcpy(&dnode->name, name);
 
     if (parent) {
     hstrcpy(&dnode->name, name);
 
     if (parent) {

-        dnode->super_block = parent->super_block;
+        vfs_d_assign_sb(dnode, parent->super_block);

         dnode->mnt = parent->mnt;
     }
 
         dnode->mnt = parent->mnt;
     }
 


@@ -395,12 +489,17 @@ vfs_d_free(struct v_dnode* dnode)
         vfs_dcache_remove(pos);
     }
 
         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*
     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;
 {
     struct hbucket* slot = &sb->i_cache[i_id & VFS_HASH_MASK];
     struct v_inode *pos, *n;


@@ -440,15 +539,15 @@ vfs_i_alloc(struct v_superblock* sb)
     memset(inode, 0, sizeof(*inode));
     mutex_init(&inode->lock);
     llist_init_head(&inode->xattrs);
     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);
 
 
     sb->ops.init_inode(sb, inode);
 

-    inode->sb = sb;

     inode->ctime = clock_unixtime();
     inode->atime = inode->ctime;
     inode->mtime = inode->ctime;
 
     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;
 }
     lru_use_one(inode_lru, &inode->lru);
     return inode;
 }


@@ -460,14 +559,28 @@ vfs_i_free(struct v_inode* inode)
         pcache_release(inode->pg_cache);
         vfree(inode->pg_cache);
     }
         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 ---- */
 
     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)
 
 int
 vfs_getfd(int fd, struct v_fd** fd_s)


@@ -478,43 +591,104 @@ vfs_getfd(int fd, struct v_fd** fd_s)
     return EBADF;
 }
 
     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,
 __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];
                       int options)
 {
     char name_str[VFS_NAME_MAXLEN];

+    struct v_dnode *fdir, *file;

     struct hstr name = HSTR(name_str, 0);
     struct hstr name = HSTR(name_str, 0);

-    int errno;
+    int errno, woption = 0;
+
+    if ((options & FLOC_NOFOLLOW)) {
+        woption |= VFS_WALK_NOFOLLOW;
+        options &= ~FLOC_NOFOLLOW;
+    }

 
 

+    floc->fresh = false;

     name_str[0] = 0;
     name_str[0] = 0;

-    if ((errno = vfs_walk_proc(path, fdir, &name, VFS_WALK_PARENT))) {
+    errno = vfs_walk_proc(path, &fdir, &name, woption | VFS_WALK_PARENT);
+    if (errno) {

         return errno;
     }
 
         return errno;
     }
 

-    errno = vfs_walk(*fdir, name.value, file, NULL, 0);
-    if (errno != ENOENT || !(options & FLOCATE_CREATE_EMPTY)) {
-        return errno;
+    errno = vfs_walk(fdir, name.value, &file, NULL, woption);
+
+    if (errno && errno != ENOENT) {
+        goto done;
+    }
+    
+    if (!errno) {
+        if ((options & FLOC_MKNAME)) {
+            errno = EEXIST;
+        }
+        goto done;

     }
 
     }
 

-    struct v_dnode* parent = *fdir;
-    struct v_dnode* file_new = vfs_d_alloc(parent, &name);
+    // errno == ENOENT
+    if (!options) {
+        goto done;
+    }

 
 

-    if (!file_new) {
-        return ENOMEM;
+    errno = vfs_check_writable(fdir);
+    if (errno) {
+        goto done;

     }
 
     }
 

-    lock_dnode(parent);
+    floc->fresh = true;

 
 

-    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);
+    file = vfs_d_alloc(fdir, &name);
+
+    if (!file) {
+        return ENOMEM;

     }
 
     }
 

-    unlock_dnode(parent);
+    lock_dnode(fdir);
+
+    vfs_dcache_add(fdir, file);
+
+done:
+    floc->dir   = fdir;
+    floc->file  = file;

 
     return errno;
 }
 
     return errno;
 }


@@ -522,29 +696,64 @@ __vfs_try_locate_file(const char* path,
 int
 vfs_do_open(const char* path, int options)
 {
 int
 vfs_do_open(const char* path, int options)
 {

-    int errno, fd;
+    int errno, fd, loptions = 0;

     struct v_dnode *dentry, *file;
     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;
     }
 
         return errno;
     }
 

-    struct v_inode* o_inode = ofile->inode;
+    file   = floc.file;
+    dentry = floc.dir;

 
 

-    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;
+    if (floc.fresh) {
+        errno = __vfs_mknod(dentry->inode, file, VFS_IFFILE, NULL);
+        if (errno) {
+            vfs_d_free(file);
+            __floc_try_unlock(&floc);
+            return errno;
+        }
+
+        __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)
 }
 
 __DEFINE_LXSYSCALL2(int, open, const char*, path, int, options)


@@ -561,13 +770,11 @@ __DEFINE_LXSYSCALL1(int, close, int, fd)
         goto done_err;
     }
 
         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;
     }
 
         goto done_err;
     }
 

-    vfree(fd_s);
+    cake_release(fd_pile, fd_s);

     __current->fdtable->fds[fd] = 0;
 
 done_err:
     __current->fdtable->fds[fd] = 0;
 
 done_err:


@@ -580,13 +787,13 @@ __vfs_readdir_callback(struct dir_context* dctx,
                        const int len,
                        const int dtype)
 {
                        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;
 }
 
     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;
 {
     struct v_fd* fd_s;
     int errno;


@@ -599,29 +806,23 @@ __DEFINE_LXSYSCALL2(int, readdir, int, fd, struct dirent*, dent)
 
     lock_inode(inode);
 
 
     lock_inode(inode);
 

-    if (!(inode->itype & VFS_IFDIR)) {
+    if (!check_directory_node(inode)) {

         errno = ENOTDIR;
         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, DT_DIR);
-        } else if (dent->d_offset == 1) {
-            __vfs_readdir_callback(&dctx, vfs_ddot.value, vfs_ddot.len, DT_DIR);
-        } 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:
     unlock_inode(inode);
 
 done:


@@ -637,7 +838,7 @@ __DEFINE_LXSYSCALL3(int, read, int, fd, void*, buf, size_t, count)
     }
 
     struct v_file* file = fd_s->file;
     }
 
     struct v_file* file = fd_s->file;

-    if ((file->inode->itype & VFS_IFDIR)) {
+    if (check_directory_node(file->inode)) {

         errno = EISDIR;
         goto done;
     }
         errno = EISDIR;
         goto done;
     }


@@ -646,7 +847,7 @@ __DEFINE_LXSYSCALL3(int, read, int, fd, void*, buf, size_t, count)
 
     file->inode->atime = clock_unixtime();
 
 
     file->inode->atime = clock_unixtime();
 

-    if ((file->inode->itype & VFS_IFSEQDEV) || (fd_s->flags & FO_DIRECT)) {
+    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);
         errno = file->ops->read(file->inode, buf, count, file->f_pos);
     } else {
         errno = pcache_read(file->inode, buf, count, file->f_pos);


@@ -672,36 +873,41 @@ __DEFINE_LXSYSCALL3(int, write, int, fd, void*, buf, size_t, count)
         goto done;
     }
 
         goto done;
     }
 

+    struct v_inode* inode;

     struct v_file* file = fd_s->file;
 
     if ((errno = vfs_check_writable(file->dnode))) {
         goto done;
     }
 
     struct v_file* file = fd_s->file;
 
     if ((errno = vfs_check_writable(file->dnode))) {
         goto done;
     }
 

-    if ((file->inode->itype & VFS_IFDIR)) {
+    if (check_directory_node(file->inode)) {

         errno = EISDIR;
         goto done;
     }
 
         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;
 
     if (errno > 0) {
         file->f_pos += errno;

-        unlock_inode(file->inode);
+        inode->fsize = MAX(inode->fsize, file->f_pos);
+
+        unlock_inode(inode);

         return errno;
     }
 
         return errno;
     }
 

-    unlock_inode(file->inode);
+    unlock_inode(inode);

 
 done:
     return DO_STATUS(errno);
 
 done:
     return DO_STATUS(errno);


@@ -716,35 +922,42 @@ __DEFINE_LXSYSCALL3(int, lseek, int, fd, int, offset, int, options)
     }
 
     struct v_file* file = fd_s->file;
     }
 
     struct v_file* file = fd_s->file;

+    struct v_inode* inode = file->inode;

 
     if (!file->ops->seek) {
         errno = ENOTSUP;
         goto done;
     }
 
 
     if (!file->ops->seek) {
         errno = ENOTSUP;
         goto done;
     }
 

-    lock_inode(file->inode);
+    lock_inode(inode);

 
     int overflow = 0;
     int fpos = file->f_pos;
 
     int overflow = 0;
     int fpos = file->f_pos;

+
+    if (vfs_get_dtype(inode->itype) == DT_DIR) {
+        options = (options != FSEEK_END) ? options : FSEEK_SET;
+    }
+    

     switch (options) {
         case FSEEK_CUR:
     switch (options) {
         case FSEEK_CUR:

-            overflow = __builtin_sadd_overflow((int)file->f_pos, offset, &fpos);
+            overflow = sadd_of((int)file->f_pos, offset, &fpos);

             break;
         case FSEEK_END:
             break;
         case FSEEK_END:

-            overflow =
-              __builtin_sadd_overflow((int)file->inode->fsize, offset, &fpos);
+            overflow = sadd_of((int)inode->fsize, offset, &fpos);

             break;
         case FSEEK_SET:
             fpos = offset;
             break;
     }
             break;
         case FSEEK_SET:
             fpos = offset;
             break;
     }

+

     if (overflow) {
         errno = EOVERFLOW;
     if (overflow) {
         errno = EOVERFLOW;

-    } else if (!(errno = file->ops->seek(file->inode, fpos))) {
-        file->f_pos = fpos;
+    }
+    else {
+        errno = file->ops->seek(file, fpos);

     }
 
     }
 

-    unlock_inode(file->inode);
+    unlock_inode(inode);

 
 done:
     return DO_STATUS(errno);
 
 done:
     return DO_STATUS(errno);


@@ -753,7 +966,7 @@ done:
 int
 vfs_get_path(struct v_dnode* dnode, char* buf, size_t size, int depth)
 {
 int
 vfs_get_path(struct v_dnode* dnode, char* buf, size_t size, int depth)
 {

-    if (!dnode || dnode->parent == dnode) {
+    if (!dnode) {

         return 0;
     }
 
         return 0;
     }
 


@@ -761,13 +974,19 @@ vfs_get_path(struct v_dnode* dnode, char* buf, size_t size, int depth)
         return ENAMETOOLONG;
     }
 
         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 >= size) {
         return len;
     }
 

-    buf[len++] = VFS_PATH_DELIM;
+    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);
 
     size_t cpy_size = MIN(dnode->name.len, size - len);
     strncpy(buf + len, dnode->name.value, cpy_size);


@@ -781,29 +1000,42 @@ vfs_readlink(struct v_dnode* dnode, char* buf, size_t size)
 {
     const char* link;
     struct v_inode* inode = dnode->inode;
 {
     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
 vfs_get_dtype(int itype)
 {

-    switch (itype) {
-        case VFS_IFDIR:
-            return DT_DIR;
-        case VFS_IFSYMLINK:
-            return DT_SYMLINK;
-        default:
-            return DT_PIPE;
+    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)
 }
 
 __DEFINE_LXSYSCALL3(int, realpathat, int, fd, char*, buf, size_t, size)


@@ -840,16 +1072,8 @@ __DEFINE_LXSYSCALL3(int, readlink, const char*, path, char*, buf, size_t, size)
     return DO_STATUS(errno);
 }
 
     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;
 {
     int errno;
     struct v_fd* fd_s;


@@ -857,6 +1081,8 @@ __DEFINE_LXSYSCALL4(int,
         goto done;
     }
 
         goto done;
     }
 

+    pathname = pathname ? pathname : "";
+

     struct v_dnode* dnode;
     if (!(errno = vfs_walk(
             fd_s->file->dnode, pathname, &dnode, NULL, VFS_WALK_NOFOLLOW))) {
     struct v_dnode* dnode;
     if (!(errno = vfs_walk(
             fd_s->file->dnode, pathname, &dnode, NULL, VFS_WALK_NOFOLLOW))) {


@@ -918,7 +1144,7 @@ __DEFINE_LXSYSCALL1(int, rmdir, const char*, pathname)
     lock_dnode(parent);
     lock_inode(parent->inode);
 
     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);
         errno = parent->inode->ops->rmdir(parent->inode, dnode);
         if (!errno) {
             vfs_dcache_remove(dnode);


@@ -946,6 +1172,11 @@ __DEFINE_LXSYSCALL1(int, mkdir, const char*, path)
         goto done;
     }
 
         goto done;
     }
 

+    if (!(errno = vfs_walk(parent, name_value, &dir, NULL, 0))) {
+        errno = EEXIST;
+        goto done;
+    }
+

     if ((errno = vfs_check_writable(parent))) {
         goto done;
     }
     if ((errno = vfs_check_writable(parent))) {
         goto done;
     }


@@ -955,16 +1186,18 @@ __DEFINE_LXSYSCALL1(int, mkdir, const char*, path)
         goto done;
     }
 
         goto done;
     }
 

+    struct v_inode* inode = parent->inode;
+

     lock_dnode(parent);
     lock_dnode(parent);

-    lock_inode(parent->inode);
+    lock_inode(inode);

 
     if ((parent->super_block->fs->types & FSTYPE_ROFS)) {
         errno = ENOTSUP;
 
     if ((parent->super_block->fs->types & FSTYPE_ROFS)) {
         errno = ENOTSUP;

-    } else if (!parent->inode->ops->mkdir) {
+    } else if (!inode->ops->mkdir) {

         errno = ENOTSUP;
         errno = ENOTSUP;

-    } else if (!(parent->inode->itype & VFS_IFDIR)) {
+    } else if (!check_directory_node(inode)) {

         errno = ENOTDIR;
         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;
     }
         vfs_dcache_add(parent, dir);
         goto cleanup;
     }


@@ -972,7 +1205,7 @@ __DEFINE_LXSYSCALL1(int, mkdir, const char*, path)
     vfs_d_free(dir);
 
 cleanup:
     vfs_d_free(dir);
 
 cleanup:

-    unlock_inode(parent->inode);
+    unlock_inode(inode);

     unlock_dnode(parent);
 done:
     return DO_STATUS(errno);
     unlock_dnode(parent);
 done:
     return DO_STATUS(errno);


@@ -996,10 +1229,8 @@ __vfs_do_unlink(struct v_dnode* dnode)
 
     if (inode->open_count) {
         errno = EBUSY;
 
     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);
         }
         if (!errno) {
             vfs_d_free(dnode);
         }


@@ -1046,18 +1277,30 @@ done:
 __DEFINE_LXSYSCALL2(int, link, const char*, oldpath, const char*, newpath)
 {
     int errno;
 __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) {
     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);
 }
 
     return DO_STATUS(errno);
 }
 


@@ -1143,58 +1386,94 @@ done:
     return DO_STATUS(errno);
 }
 
     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;
 {
     int errno;

-    struct v_dnode* dnode;
-    if ((errno = vfs_walk_proc(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;
+    }
+
+    file = floc.file;
+    errno = __vfs_mknod(floc.dir->inode, file, VFS_IFSYMLINK, NULL);
+    if (errno) {
+        vfs_d_free(file);

         goto done;
     }
 
         goto done;
     }
 

-    if (errno = vfs_check_writable(dnode)) {
+    f_ino = file->inode;
+
+    assert(f_ino);
+
+    errno = vfs_check_writable(file);
+    if (errno) {

         goto done;
     }
 
         goto done;
     }
 

-    if (!dnode->inode->ops->set_symlink) {
+    if (!f_ino->ops->set_symlink) {

         errno = ENOTSUP;
         goto done;
     }
 
         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:
 
 done:

+    __floc_try_unlock(&floc);

     return DO_STATUS(errno);
 }
 
     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
 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);
 
 {
     int errno = 0;
 
     lock_dnode(dnode);
 

-    if (!(dnode->inode->itype & VFS_IFDIR)) {
+    if (!check_directory_node(dnode->inode)) {

         errno = ENOTDIR;
         goto done;
     }
 
         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);
 
 
     unlock_dnode(dnode);
 


@@ -1211,7 +1490,7 @@ __DEFINE_LXSYSCALL1(int, chdir, const char*, path)
         goto done;
     }
 
         goto done;
     }
 

-    errno = __vfs_do_chdir(dnode);
+    errno = vfs_do_chdir((struct proc_info*)__current, dnode);

 
 done:
     return DO_STATUS(errno);
 
 done:
     return DO_STATUS(errno);


@@ -1226,7 +1505,7 @@ __DEFINE_LXSYSCALL1(int, fchdir, int, fd)
         goto done;
     }
 
         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);
 
 done:
     return DO_STATUS(errno);


@@ -1254,12 +1533,12 @@ __DEFINE_LXSYSCALL2(char*, getcwd, char*, buf, size_t, size)
         }
     }
 
         }
     }
 

-    buf[len + 1] = '\0';
+    buf[len] = '\0';

 
     ret_ptr = buf;
 
 done:
 
     ret_ptr = buf;
 
 done:

-    __current->k_status = errno;
+    syscall_result(errno);

     return ret_ptr;
 }
 
     return ret_ptr;
 }
 


@@ -1272,7 +1551,7 @@ vfs_do_rename(struct v_dnode* current, struct v_dnode* target)
         return 0;
     }
 
         return 0;
     }
 

-    if (errno = vfs_check_writable(current)) {
+    if ((errno = vfs_check_writable(current))) {

         return errno;
     }
 
         return errno;
     }
 


@@ -1356,6 +1635,47 @@ __DEFINE_LXSYSCALL2(int, rename, const char*, oldpath, const char*, newpath)
     errno = vfs_do_rename(cur, target);
 
 done:
     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 || rdev->magic != DEV_STRUCT_MAGIC) {
+            errno = EINVAL;
+            goto done;
+        }
+
+        stat->st_rdev = (dev_t){.meta = rdev->ident.fn_grp,
+                                .unique = rdev->ident.unique,
+                                .index = rdev->dev_uid};
+    }
+
+    if (fdev) {
+        stat->st_dev = (dev_t){.meta = fdev->ident.fn_grp,
+                               .unique = fdev->ident.unique,
+                               .index = fdev->dev_uid};
+    }
+
+done:

     return DO_STATUS(errno);
 }
\ No newline at end of file
     return DO_STATUS(errno);
 }
\ No newline at end of file