vfs_intent-2.6-rhel4.patch

非常经典的一个分布式系统

diff -rup RH_2_6_9_55.orig/fs/cifs/dir.c RH_2_6_9_55/fs/cifs/dir.c
--- RH_2_6_9_55.orig/fs/cifs/dir.c
+++ RH_2_6_9_55/fs/cifs/dir.c
@@ -157,11 +157,7 @@ cifs_create(struct inode *inode, struct 

 #if LINUX_VERSION_CODE > KERNEL_VERSION(2, 5, 0)
	if(nd && (nd->flags & LOOKUP_OPEN)) {
-#if LINUX_VERSION_CODE == KERNEL_VERSION(2,6,5) /* SUSE included Lustre patch */
		int oflags = nd->intent.it_flags;
-#else
-		int oflags = nd->intent.open.flags;
-#endif

		desiredAccess = 0;
		if (oflags & FMODE_READ)
diff -rup RH_2_6_9_55.orig/fs/exec.c RH_2_6_9_55/fs/exec.c
--- RH_2_6_9_55.orig/fs/exec.c
+++ RH_2_6_9_55/fs/exec.c
@@ -126,9 +126,10 @@ asmlinkage long sys_uselib(const char __
 	struct file * file;
 	struct nameidata nd;
 	int error;
-
-	nd.intent.open.flags = FMODE_READ|FMODE_EXEC;
-	error = __user_walk(library, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
+	intent_init(&nd.intent, IT_OPEN);
+  
+	nd.intent.it_flags = FMODE_READ|FMODE_EXEC;
+	error = __user_walk_it(library, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd);
 	if (error)
 		goto out;
 
@@ -140,7 +141,7 @@ asmlinkage long sys_uselib(const char __
 	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;
@@ -489,8 +490,9 @@ struct file *open_exec(const char *name)
 	int err;
 	struct file *file;
 
-	nd.intent.open.flags = FMODE_READ|FMODE_EXEC;
-	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_it(name, LOOKUP_FOLLOW, &nd);
 	file = ERR_PTR(err);
 
 	if (!err) {
@@ -501,7 +503,7 @@ struct file *open_exec(const char *name)
			int err = permission(inode, MAY_EXEC, &nd);
 			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) {
diff -rup RH_2_6_9_55.orig/fs/inode.c RH_2_6_9_55/fs/inode.c
--- RH_2_6_9_55.orig/fs/inode.c
+++ RH_2_6_9_55/fs/inode.c
@@ -235,6 +235,7 @@ void __iget(struct inode * inode)
 	inodes_stat.nr_unused--;
 }
 
+EXPORT_SYMBOL(__iget);
 /**
  * clear_inode - clear an inode
  * @inode: inode to clear
diff -rup RH_2_6_9_55.orig/fs/namei.c RH_2_6_9_55/fs/namei.c
--- RH_2_6_9_55.orig/fs/namei.c
+++ RH_2_6_9_55/fs/namei.c
@@ -282,8 +282,19 @@ int deny_write_access(struct file * file
 	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);
 }
@@ -395,8 +406,12 @@ static struct dentry * real_lookup(struc
 {
 	struct dentry * result;
 	struct inode *dir = parent->d_inode;
-
+	int counter = 0;
+	
+again:
 	down(&dir->i_sem);
+	counter++;
+
 	/*
 	 * First re-do the cached lookup just in case it was created
	 * while we waited for the directory semaphore..
@@ -433,8 +448,12 @@ static struct dentry * real_lookup(struc
	up(&dir->i_sem);
 	if (result->d_op && result->d_op->d_revalidate) {
		result = do_revalidate(result, nd);
-		if (!result)
-			result = ERR_PTR(-ENOENT);
+		if (!result) {
+			if (counter > 10)
+				result = ERR_PTR(-ESTALE);
+			if (!IS_ERR(result))
+				goto again;
+		}
 	}
 	return result;
 }
@@ -464,6 +483,7 @@ static inline int __vfs_follow_link(stru
 {
 	int res = 0;
 	char *name;
+
 	if (IS_ERR(link))
 		goto fail;
 
@@ -473,6 +493,7 @@ static inline int __vfs_follow_link(stru
 			/* 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)
@@ -681,6 +702,33 @@ fail:
 	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.
  * This is the basic name resolution function, turning a pathname into
@@ -782,13 +830,17 @@ static fastcall int __link_path_walk(con
 			goto out_dput;
 
 		if (inode->i_op->follow_link) {
+			int save_flags = nd->flags;
 			mntget(next.mnt);
			if (next.mnt != nd->mnt) {
				dput(nd->dentry);
				nd->mnt = next.mnt;
				nd->dentry = dget(next.dentry);
			}
+			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)
@@ -828,14 +880,34 @@ last_component:
 				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);
@@ -1007,7 +1079,7 @@ set_it:
 }
 
 /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */
-int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata *nd)
+int fastcall path_lookup_it(const char *name, unsigned int flags, struct nameidata *nd)
 {
 	int retval = 0;
 
@@ -1041,6 +1113,12 @@ out:
 	return retval;
 }
 
+int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata *nd)
+{
+	intent_init(&nd->intent, IT_GETATTR);
+	return path_lookup_it(name, flags, nd);
+}
+
 /*
  * Restricted form of lookup. Doesn't follow links, single-component only,
  * needs parent already locked. Doesn't follow mounts.
@@ -1091,7 +1169,7 @@ struct dentry * lookup_hash(struct qstr 
 }
 
 /* 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;
@@ -1111,11 +1189,16 @@ struct dentry * lookup_one_len(const cha
 	}
 	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()
  *
@@ -1127,18 +1210,24 @@ access:
  * 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);
 
 	if (!IS_ERR(tmp)) {
-		err = path_lookup(tmp, flags, nd);
+		err = path_lookup_it(tmp, flags, nd);
 		putname(tmp);
 	}
 	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.
@@ -1384,7 +1473,7 @@ int may_open(struct nameidata *nd, int a
 		if (!error) {
 			DQUOT_INIT(inode);
 			
-			error = do_truncate(dentry, 0, ATTR_MTIME|ATTR_CTIME);
+			error = do_truncate(dentry, 0, ATTR_MTIME|ATTR_CTIME|ATTR_FROM_OPEN);
 		}
 		put_write_access(inode);
 		if (error)
@@ -1425,14 +1514,14 @@ int open_namei(const char * pathname, in
 		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.
 	 */
 	if (!(flag & O_CREAT)) {
-		error = path_lookup(pathname, lookup_flags(flag)|LOOKUP_OPEN, nd);
+		error = path_lookup_it(pathname, lookup_flags(flag)|LOOKUP_OPEN, nd);
 		if (error)
 			return error;
 		goto ok;
@@ -1441,7 +1530,8 @@ int open_namei(const char * pathname, in
 	/*
 	 * Create - we need to know the parent.
 	 */
-	error = path_lookup(pathname, LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE, nd);
+	nd->intent.it_op |= IT_CREAT;
+	error = path_lookup_it(pathname, LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE, nd);
 	if (error)
 		return error;
 
@@ -1457,7 +1547,9 @@ int open_namei(const char * pathname, in
 	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);
@@ -1570,7 +1662,9 @@ do_link:
 	}
 	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;
 }
@@ -1644,10 +1738,20 @@ asmlinkage long sys_mknod(const char __u
 	tmp = getname(filename);
 	if (IS_ERR(tmp))
 		return PTR_ERR(tmp);
-
-	error = path_lookup(tmp, LOOKUP_PARENT, &nd);
+		
+	intent_init(&nd.intent, IT_LOOKUP);
+	error = path_lookup_it(tmp, LOOKUP_PARENT, &nd);
 	if (error)
 		goto out;
+
+	if (nd.dentry->d_inode->i_op->mknod_raw) {
+		struct inode_operations *op = nd.dentry->d_inode->i_op;
+		error = op->mknod_raw(&nd, mode, dev);
+		/* the file system wants to use normal vfs path now */
+		if (error != -EOPNOTSUPP)
+			goto out2;
+	}
+
 	dentry = lookup_create(&nd, 0);
 	error = PTR_ERR(dentry);
 
@@ -1674,6 +1778,7 @@ asmlinkage long sys_mknod(const char __u
 		dput(dentry);
 	}
 	up(&nd.dentry->d_inode->i_sem);
+out2:
 	path_release(&nd);
 out:
 	putname(tmp);
@@ -1716,10 +1821,20 @@ asmlinkage long sys_mkdir(const char __u
 	if (!IS_ERR(tmp)) {
 		struct dentry *dentry;
 		struct nameidata nd;
+		intent_init(&nd.intent, IT_LOOKUP);
 
-		error = path_lookup(tmp, LOOKUP_PARENT, &nd);
+		error = path_lookup_it(tmp, LOOKUP_PARENT, &nd);
 		if (error)
 			goto out;
+
+		if (nd.dentry->d_inode->i_op->mkdir_raw) {
+			struct inode_operations *op = nd.dentry->d_inode->i_op;
+			error = op->mkdir_raw(&nd, mode);
+			/* the file system wants to use normal vfs path now */
+			if (error != -EOPNOTSUPP)
+				goto out2;
+		}
+
 		dentry = lookup_create(&nd, 1);
 		error = PTR_ERR(dentry);
 		if (!IS_ERR(dentry)) {
@@ -1729,6 +1844,7 @@ asmlinkage long sys_mkdir(const char __u
 			dput(dentry);
 		}
 		up(&nd.dentry->d_inode->i_sem);
+out2:
 		path_release(&nd);
 out:
 		putname(tmp);
@@ -1814,7 +1930,8 @@ asmlinkage long sys_rmdir(const char __u
 	if(IS_ERR(name))
 		return PTR_ERR(name);
 
-	error = path_lookup(name, LOOKUP_PARENT, &nd);
+	intent_init(&nd.intent, IT_LOOKUP);
+	error = path_lookup_it(name, LOOKUP_PARENT, &nd);
 	if (error)
 		goto exit;
 
@@ -1829,6 +1946,16 @@ asmlinkage long sys_rmdir(const char __u
 			error = -EBUSY;
 			goto exit1;
 	}
+
+	if (nd.dentry->d_inode->i_op->rmdir_raw) {
+		struct inode_operations *op = nd.dentry->d_inode->i_op;
+
+		error = op->rmdir_raw(&nd);
+		/* the file system wants to use normal vfs path now */
+		if (error != -EOPNOTSUPP)
+			goto exit1;
+	}
+
 	down(&nd.dentry->d_inode->i_sem);
 	dentry = lookup_hash(&nd.last, nd.dentry);
 	error = PTR_ERR(dentry);
@@ -1892,12 +2019,22 @@ asmlinkage long sys_unlink(const char __
 	if(IS_ERR(name))
 		return PTR_ERR(name);
 
-	error = path_lookup(name, LOOKUP_PARENT, &nd);
+	intent_init(&nd.intent, IT_LOOKUP);
+	error = path_lookup_it(name, LOOKUP_PARENT, &nd);
 	if (error)
 		goto exit;
 	error = -EISDIR;
 	if (nd.last_type != LAST_NORM)
 		goto exit1;
+	
+	if (nd.dentry->d_inode->i_op->unlink_raw) {
+		struct inode_operations *op = nd.dentry->d_inode->i_op;
+		error = op->unlink_raw(&nd);
+		/* the file system wants to use normal vfs path now */
+		if (error != -EOPNOTSUPP)
+			goto exit1;
+	}
+
 	down(&nd.dentry->d_inode->i_sem);
 	dentry = lookup_hash(&nd.last, nd.dentry);
 	error = PTR_ERR(dentry);
@@ -1965,10 +2102,20 @@ asmlinkage long sys_symlink(const char _
 	if (!IS_ERR(to)) {
 		struct dentry *dentry;
 		struct nameidata nd;
+		intent_init(&nd.intent, IT_LOOKUP);
 
-		error = path_lookup(to, LOOKUP_PARENT, &nd);
+		error = path_lookup_it(to, LOOKUP_PARENT, &nd);
 		if (error)
 			goto out;