vfs_intent-2.6-fc3.patch

非常经典的一个分布式系统

--- linux-2.6.10.orig/fs/exec.c
+++ linux-2.6.10/fs/exec.c
@@ -124,9 +124,10 @@
 	struct file * file;
 	struct nameidata nd;
 	int error;
+	intent_init(&nd.intent, IT_OPEN);
 
-	nd.intent.open.flags = FMODE_READ;
-	error = __user_walk(library, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
+	nd.intent.it_flags = FMODE_READ|FMODE_EXEC;
+	error = __user_walk_it(library, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
 	if (error)
 		goto out;
 
@@ -138,7 +139,7 @@
 	if (error)
 		goto exit;
 
-	file = dentry_open(nd.dentry, nd.mnt, O_RDONLY);
+	file = dentry_open_it(nd.dentry, nd.mnt, O_RDONLY, &nd.intent);
 	error = PTR_ERR(file);
 	if (IS_ERR(file))
 		goto out;
@@ -485,8 +486,9 @@
 	int err;
 	struct file *file;
 
-	nd.intent.open.flags = FMODE_READ;
-	err = path_lookup(name, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
+	intent_init(&nd.intent, IT_OPEN);
+	nd.intent.it_flags = FMODE_READ|FMODE_EXEC;
+	err = path_lookup(name, LOOKUP_FOLLOW, &nd);
 	file = ERR_PTR(err);
 
 	if (!err) {
@@ -499,7 +501,7 @@
 				err = -EACCES;
 			file = ERR_PTR(err);
 			if (!err) {
-				file = dentry_open(nd.dentry, nd.mnt, O_RDONLY);
+				file = dentry_open_it(nd.dentry, nd.mnt, O_RDONLY, &nd.intent);
 				if (!IS_ERR(file)) {
 					err = deny_write_access(file);
 					if (err) {
--- linux-2.6.10.orig/fs/inode.c
+++ linux-2.6.10/fs/inode.c
@@ -233,6 +233,7 @@
 	inodes_stat.nr_unused--;
 }
 
+EXPORT_SYMBOL(__iget);
 /**
  * clear_inode - clear an inode
  * @inode: inode to clear
--- linux-2.6.10.orig/fs/namei.c
+++ linux-2.6.10/fs/namei.c
@@ -288,8 +288,19 @@
 	return 0;
 }
 
+void intent_release(struct lookup_intent *it)
+{
+	if (!it)
+		return;
+	if (it->it_magic != INTENT_MAGIC)
+		return;
+	if (it->it_op_release)
+		it->it_op_release(it);
+}
+
 void path_release(struct nameidata *nd)
 {
+	intent_release(&nd->intent);
 	dput(nd->dentry);
 	mntput(nd->mnt);
 }
@@ -379,7 +390,10 @@
 {
 	struct dentry * result;
 	struct inode *dir = parent->d_inode;
+	int counter = 0;
 
+again:
+	counter++;
 	down(&dir->i_sem);
 	/*
 	 * First re-do the cached lookup just in case it was created
@@ -418,7 +432,10 @@
 	if (result->d_op && result->d_op->d_revalidate) {
 		if (!result->d_op->d_revalidate(result, nd) && !d_invalidate(result)) {
 			dput(result);
-			result = ERR_PTR(-ENOENT);
+			if (counter > 10)
+				result = ERR_PTR(-ESTALE);
+			if (!IS_ERR(result))
+				goto again;
 		}
 	}
 	return result;
@@ -449,6 +466,7 @@
 {
 	int res = 0;
 	char *name;
+
 	if (IS_ERR(link))
 		goto fail;
 
@@ -458,6 +476,7 @@
 			/* weird __emul_prefix() stuff did it */
 			goto out;
 	}
+	intent_reset_fs_part(&nd->intent);
 	res = link_path_walk(link, nd);
 out:
 	if (nd->depth || res || nd->last_type!=LAST_NORM)
@@ -666,6 +685,33 @@
 	return PTR_ERR(dentry);
 }
 
+static int revalidate_special(struct nameidata *nd)
+{
+	struct dentry *dentry = nd->dentry;
+	int err, counter = 0;
+
+ revalidate_again:
+	if (!dentry->d_op || !dentry->d_op->d_revalidate)
+		return 0;
+	if (!dentry->d_op->d_revalidate(dentry, nd)) {
+		struct dentry *new;
+		if ((err = permission(dentry->d_parent->d_inode, MAY_EXEC, nd)))
+			return err;
+		new = real_lookup(dentry->d_parent, &dentry->d_name, nd);
+		if (IS_ERR(new))
+			return PTR_ERR(new);
+		d_invalidate(dentry);
+		dput(dentry);
+		nd->dentry = dentry = new;
+		counter++;
+		if (counter < 10)
+			goto revalidate_again;
+		printk("excessive revalidate_it loops\n");
+		return -ESTALE;
+	}
+	return 0;
+}
+
 /*
  * Name resolution.
  *
@@ -767,8 +813,12 @@
 			goto out_dput;
 
 		if (inode->i_op->follow_link) {
+			int save_flags = nd->flags;
 			mntget(next.mnt);
+			nd->flags |= LOOKUP_LINK_NOTLAST;
 			err = do_follow_link(next.dentry, nd);
+			if (!(save_flags & LOOKUP_LINK_NOTLAST))
+				nd->flags &= ~LOOKUP_LINK_NOTLAST;
 			dput(next.dentry);
 			mntput(next.mnt);
 			if (err)
@@ -807,14 +857,34 @@
 				inode = nd->dentry->d_inode;
 				/* fallthrough */
 			case 1:
+ 				nd->flags |= LOOKUP_LAST;
+ 				err = revalidate_special(nd);
+ 				nd->flags &= ~LOOKUP_LAST;
+				if (!nd->dentry->d_inode)
+					err = -ENOENT;
+				if (err) {
+ 					path_release(nd);
+					goto return_err;
+				}
+				if (lookup_flags & LOOKUP_DIRECTORY) {
+					err = -ENOTDIR;
+					if (!nd->dentry->d_inode->i_op ||
+					    !nd->dentry->d_inode->i_op->lookup){
+						path_release(nd);
+						goto return_err;
+					}
+				}
 				goto return_reval;
 		}
+
 		if (nd->dentry->d_op && nd->dentry->d_op->d_hash) {
 			err = nd->dentry->d_op->d_hash(nd->dentry, &this);
 			if (err < 0)
 				break;
 		}
+		nd->flags |= LOOKUP_LAST;
 		err = do_lookup(nd, &this, &next, atomic);
+		nd->flags &= ~LOOKUP_LAST;
 		if (err)
 			break;
 		follow_mount(&next.mnt, &next.dentry);
@@ -1032,7 +1102,7 @@
 }
 
 /* SMP-safe */
-struct dentry * lookup_one_len(const char * name, struct dentry * base, int len)
+struct dentry * lookup_one_len_it(const char * name, struct dentry * base, int len, struct nameidata *nd)
 {
 	unsigned long hash;
 	struct qstr this;
@@ -1052,11 +1122,16 @@
 	}
 	this.hash = end_name_hash(hash);
 
-	return lookup_hash(&this, base);
+	return __lookup_hash(&this, base, nd);
 access:
 	return ERR_PTR(-EACCES);
 }
 
+struct dentry * lookup_one_len(const char * name, struct dentry * base, int len)
+{
+	return lookup_one_len_it(name, base, len, NULL);
+}
+
 /*
  *	namei()
  *
@@ -1068,7 +1143,7 @@
  * that namei follows links, while lnamei does not.
  * SMP-safe
  */
-int fastcall __user_walk(const char __user *name, unsigned flags, struct nameidata *nd)
+int fastcall __user_walk_it(const char __user *name, unsigned flags, struct nameidata *nd)
 {
 	char *tmp = getname(name);
 	int err = PTR_ERR(tmp);
@@ -1080,6 +1155,12 @@
 	return err;
 }
 
+int fastcall __user_walk(const char __user *name, unsigned flags, struct nameidata *nd)
+{
+	intent_init(&nd->intent, IT_LOOKUP);
+	return __user_walk_it(name, flags, nd);
+}
+
 /*
  * It's inline, so penalty for filesystems that don't use sticky bit is
  * minimal.
@@ -1363,8 +1444,8 @@
 		acc_mode |= MAY_APPEND;
 
 	/* Fill in the open() intent data */
-	nd->intent.open.flags = flag;
-	nd->intent.open.create_mode = mode;
+	nd->intent.it_flags = flag;
+	nd->intent.it_create_mode = mode;
 
 	/*
 	 * The simplest case - just a plain lookup.
@@ -1379,6 +1460,7 @@
 	/*
 	 * Create - we need to know the parent.
 	 */
+	nd->intent.it_op |= IT_CREAT;
 	error = path_lookup(pathname, LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE, nd);
 	if (error)
 		return error;
@@ -1395,7 +1477,9 @@
 	dir = nd->dentry;
 	nd->flags &= ~LOOKUP_PARENT;
 	down(&dir->d_inode->i_sem);
+	nd->flags |= LOOKUP_LAST;
 	dentry = __lookup_hash(&nd->last, nd->dentry, nd);
+	nd->flags &= ~LOOKUP_LAST;
 
 do_last:
 	error = PTR_ERR(dentry);
@@ -1508,7 +1592,9 @@
 	}
 	dir = nd->dentry;
 	down(&dir->d_inode->i_sem);
+	nd->flags |= LOOKUP_LAST;
 	dentry = __lookup_hash(&nd->last, nd->dentry, nd);
+	nd->flags &= ~LOOKUP_LAST;
 	putname(nd->last.name);
 	goto do_last;
 }
--- linux-2.6.10.orig/fs/namespace.c
+++ linux-2.6.10/fs/namespace.c
@@ -62,6 +62,7 @@
 		INIT_LIST_HEAD(&mnt->mnt_mounts);
 		INIT_LIST_HEAD(&mnt->mnt_list);
 		INIT_LIST_HEAD(&mnt->mnt_fslink);
+		INIT_LIST_HEAD(&mnt->mnt_lustre_list);
 		if (name) {
 			int size = strlen(name)+1;
 			char *newname = kmalloc(size, GFP_KERNEL);
@@ -113,6 +114,7 @@
 
 static void detach_mnt(struct vfsmount *mnt, struct nameidata *old_nd)
 {
+	memset(old_nd, 0, sizeof(*old_nd));
 	old_nd->dentry = mnt->mnt_mountpoint;
 	old_nd->mnt = mnt->mnt_parent;
 	mnt->mnt_parent = mnt;
@@ -176,6 +178,9 @@
 {
 	struct super_block *sb = mnt->mnt_sb;
 	dput(mnt->mnt_root);
+	spin_lock(&dcache_lock);
+	list_del(&mnt->mnt_lustre_list);
+	spin_unlock(&dcache_lock);
 	free_vfsmnt(mnt);
 	deactivate_super(sb);
 }
@@ -402,6 +407,8 @@
 	 */
 
 	lock_kernel();
+	if (sb->s_op->umount_lustre)
+		sb->s_op->umount_lustre(sb);
 	if( (flags&MNT_FORCE) && sb->s_op->umount_begin)
 		sb->s_op->umount_begin(sb);
 	unlock_kernel();
@@ -627,6 +634,7 @@
 		return err;
 	if (!old_name || !*old_name)
 		return -EINVAL;
+	intent_init(&old_nd.intent, IT_LOOKUP);
 	err = path_lookup(old_name, LOOKUP_FOLLOW, &old_nd);
 	if (err)
 		return err;
@@ -701,6 +709,7 @@
 		return -EPERM;
 	if (!old_name || !*old_name)
 		return -EINVAL;
+	intent_init(&old_nd.intent, IT_LOOKUP);
 	err = path_lookup(old_name, LOOKUP_FOLLOW, &old_nd);
 	if (err)
 		return err;
@@ -1012,6 +1021,7 @@
 	int retval = 0;
 	int mnt_flags = 0;
 
+	intent_init(&nd.intent, IT_LOOKUP);
 	/* Discard magic */
 	if ((flags & MS_MGC_MSK) == MS_MGC_VAL)
 		flags &= ~MS_MGC_MSK;
--- linux-2.6.10.orig/fs/open.c
+++ linux-2.6.10/fs/open.c
@@ -216,12 +216,12 @@
 	struct nameidata nd;
 	struct inode * inode;
 	int error;
-
+	intent_init(&nd.intent, IT_GETATTR);
 	error = -EINVAL;
 	if (length < 0)	/* sorry, but loff_t says... */
 		goto out;
 
-	error = user_path_walk(path, &nd);
+	error = user_path_walk_it(path, &nd);
 	if (error)
 		goto out;
 	inode = nd.dentry->d_inode;
@@ -475,6 +475,7 @@
 	int old_fsuid, old_fsgid;
 	kernel_cap_t old_cap;
 	int res;
+	intent_init(&nd.intent, IT_GETATTR);
 
 	if (mode & ~S_IRWXO)	/* where's F_OK, X_OK, W_OK, R_OK? */
 		return -EINVAL;
@@ -499,13 +500,14 @@
 	else
 		current->cap_effective = current->cap_permitted;
 
-	res = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd);
+	res = __user_walk_it(filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd);
 	if (!res) {