configfs lookup race fix

configfs lookup race fix

[Christoph Hellwig]

Hi all,

this series takes the patch from Sishuai, with the initial refactoring
split into reviewable prep patches and the suggestion from Al taken
into account.  It does not fix the pre-existing leak of ->s_dentry and
->d_fsdata that Al noticed yet - that will take a little more time.

[PATCH 1/4] configfs: return -ENAMETOOLONG earlier in configfs_lookup

[Christoph Hellwig]

Just like most other file systems: get the simple checks out of the
way first.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/configfs/dir.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index ac5e0c0e9181..cf08bbde55f3 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -456,6 +456,9 @@ static struct dentry * configfs_lookup(struct inode *dir,
 	int found = 0;
 	int err;
 
+	if (dentry->d_name.len > NAME_MAX)
+		return ERR_PTR(-ENAMETOOLONG);
+
 	/*
 	 * Fake invisibility if dir belongs to a group/default groups hierarchy
 	 * being attached
@@ -486,8 +489,6 @@ static struct dentry * configfs_lookup(struct inode *dir,
 		 * If it doesn't exist and it isn't a NOT_PINNED item,
 		 * it must be negative.
 		 */
-		if (dentry->d_name.len > NAME_MAX)
-			return ERR_PTR(-ENAMETOOLONG);
 		d_add(dentry, NULL);
 		return NULL;
 	}
-- 
2.30.2

Re: [PATCH 1/4] configfs: return -ENAMETOOLONG earlier in configfs_lookup

[Joel Becker]

Reviewed-by: Joel Becker <jlbec@evilplan.org>

On Wed, Aug 25, 2021 at 08:49:03AM +0200, Christoph Hellwig wrote:
> Just like most other file systems: get the simple checks out of the
> way first.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/configfs/dir.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
> index ac5e0c0e9181..cf08bbde55f3 100644
> --- a/fs/configfs/dir.c
> +++ b/fs/configfs/dir.c
> @@ -456,6 +456,9 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  	int found = 0;
>  	int err;
>  
> +	if (dentry->d_name.len > NAME_MAX)
> +		return ERR_PTR(-ENAMETOOLONG);
> +
>  	/*
>  	 * Fake invisibility if dir belongs to a group/default groups hierarchy
>  	 * being attached
> @@ -486,8 +489,6 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  		 * If it doesn't exist and it isn't a NOT_PINNED item,
>  		 * it must be negative.
>  		 */
> -		if (dentry->d_name.len > NAME_MAX)
> -			return ERR_PTR(-ENAMETOOLONG);
>  		d_add(dentry, NULL);
>  		return NULL;
>  	}
> -- 
> 2.30.2
> 

-- 

Re: [PATCH 1/4] configfs: return -ENAMETOOLONG earlier in configfs_lookup

[Joel Becker]

Acked-by: Joel Becker <jlbec@evilplan.org>

On Wed, Aug 25, 2021 at 08:49:03AM +0200, Christoph Hellwig wrote:
> Just like most other file systems: get the simple checks out of the
> way first.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/configfs/dir.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
> index ac5e0c0e9181..cf08bbde55f3 100644
> --- a/fs/configfs/dir.c
> +++ b/fs/configfs/dir.c
> @@ -456,6 +456,9 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  	int found = 0;
>  	int err;
>  
> +	if (dentry->d_name.len > NAME_MAX)
> +		return ERR_PTR(-ENAMETOOLONG);
> +
>  	/*
>  	 * Fake invisibility if dir belongs to a group/default groups hierarchy
>  	 * being attached
> @@ -486,8 +489,6 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  		 * If it doesn't exist and it isn't a NOT_PINNED item,
>  		 * it must be negative.
>  		 */
> -		if (dentry->d_name.len > NAME_MAX)
> -			return ERR_PTR(-ENAMETOOLONG);
>  		d_add(dentry, NULL);
>  		return NULL;
>  	}
> -- 
> 2.30.2
> 

-- 

[PATCH 2/4] configfs: simplify the configfs_dirent_is_ready

[Christoph Hellwig]

Return the error directly instead of using a goto.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/configfs/dir.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index cf08bbde55f3..5d58569f0eea 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -467,9 +467,8 @@ static struct dentry * configfs_lookup(struct inode *dir,
 	 * not complete their initialization, since the dentries of the
 	 * attributes won't be instantiated.
 	 */
-	err = -ENOENT;
 	if (!configfs_dirent_is_ready(parent_sd))
-		goto out;
+		return ERR_PTR(-ENOENT);
 
 	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
 		if (sd->s_type & CONFIGFS_NOT_PINNED) {
@@ -493,7 +492,6 @@ static struct dentry * configfs_lookup(struct inode *dir,
 		return NULL;
 	}
 
-out:
 	return ERR_PTR(err);
 }
 
-- 
2.30.2

Re: [PATCH 2/4] configfs: simplify the configfs_dirent_is_ready

[Joel Becker]

Acked-by: Joel Becker <jlbec@evilplan.org>

On Wed, Aug 25, 2021 at 08:49:04AM +0200, Christoph Hellwig wrote:
> Return the error directly instead of using a goto.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/configfs/dir.c | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
> index cf08bbde55f3..5d58569f0eea 100644
> --- a/fs/configfs/dir.c
> +++ b/fs/configfs/dir.c
> @@ -467,9 +467,8 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  	 * not complete their initialization, since the dentries of the
>  	 * attributes won't be instantiated.
>  	 */
> -	err = -ENOENT;
>  	if (!configfs_dirent_is_ready(parent_sd))
> -		goto out;
> +		return ERR_PTR(-ENOENT);
>  
>  	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
>  		if (sd->s_type & CONFIGFS_NOT_PINNED) {
> @@ -493,7 +492,6 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  		return NULL;
>  	}
>  
> -out:
>  	return ERR_PTR(err);
>  }
>  
> -- 
> 2.30.2
> 

-- 

[PATCH 3/4] configfs: fold configfs_attach_attr into configfs_lookup

[Christoph Hellwig]

This makes it more clear what gets added to the dcache and prepares
for an additional locking fix.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/configfs/dir.c | 73 ++++++++++++++++-------------------------------
 1 file changed, 24 insertions(+), 49 deletions(-)

diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 5d58569f0eea..fc20bd8a6337 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -45,7 +45,7 @@ static void configfs_d_iput(struct dentry * dentry,
 		/*
 		 * Set sd->s_dentry to null only when this dentry is the one
 		 * that is going to be killed.  Otherwise configfs_d_iput may
-		 * run just after configfs_attach_attr and set sd->s_dentry to
+		 * run just after configfs_lookup and set sd->s_dentry to
 		 * NULL even it's still in use.
 		 */
 		if (sd->s_dentry == dentry)
@@ -417,44 +417,13 @@ static void configfs_remove_dir(struct config_item * item)
 	dput(dentry);
 }
 
-
-/* attaches attribute's configfs_dirent to the dentry corresponding to the
- * attribute file
- */
-static int configfs_attach_attr(struct configfs_dirent * sd, struct dentry * dentry)
-{
-	struct configfs_attribute * attr = sd->s_element;
-	struct inode *inode;
-
-	spin_lock(&configfs_dirent_lock);
-	dentry->d_fsdata = configfs_get(sd);
-	sd->s_dentry = dentry;
-	spin_unlock(&configfs_dirent_lock);
-
-	inode = configfs_create(dentry, (attr->ca_mode & S_IALLUGO) | S_IFREG);
-	if (IS_ERR(inode)) {
-		configfs_put(sd);
-		return PTR_ERR(inode);
-	}
-	if (sd->s_type & CONFIGFS_ITEM_BIN_ATTR) {
-		inode->i_size = 0;
-		inode->i_fop = &configfs_bin_file_operations;
-	} else {
-		inode->i_size = PAGE_SIZE;
-		inode->i_fop = &configfs_file_operations;
-	}
-	d_add(dentry, inode);
-	return 0;
-}
-
 static struct dentry * configfs_lookup(struct inode *dir,
 				       struct dentry *dentry,
 				       unsigned int flags)
 {
 	struct configfs_dirent * parent_sd = dentry->d_parent->d_fsdata;
 	struct configfs_dirent * sd;
-	int found = 0;
-	int err;
+	struct inode *inode = NULL;
 
 	if (dentry->d_name.len > NAME_MAX)
 		return ERR_PTR(-ENAMETOOLONG);
@@ -471,28 +440,34 @@ static struct dentry * configfs_lookup(struct inode *dir,
 		return ERR_PTR(-ENOENT);
 
 	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
-		if (sd->s_type & CONFIGFS_NOT_PINNED) {
-			const unsigned char * name = configfs_get_name(sd);
+		if ((sd->s_type & CONFIGFS_NOT_PINNED) &&
+		    !strcmp(configfs_get_name(sd), dentry->d_name.name)) {
+			struct configfs_attribute *attr = sd->s_element;
+			umode_t mode = (attr->ca_mode & S_IALLUGO) | S_IFREG;
 
-			if (strcmp(name, dentry->d_name.name))
-				continue;
+			spin_lock(&configfs_dirent_lock);
+			dentry->d_fsdata = configfs_get(sd);
+			sd->s_dentry = dentry;
+			spin_unlock(&configfs_dirent_lock);
 
-			found = 1;
-			err = configfs_attach_attr(sd, dentry);
+			inode = configfs_create(dentry, mode);
+			if (IS_ERR(inode)) {
+				configfs_put(sd);
+				return ERR_CAST(inode);
+			}
+			if (sd->s_type & CONFIGFS_ITEM_BIN_ATTR) {
+				inode->i_size = 0;
+				inode->i_fop = &configfs_bin_file_operations;
+			} else {
+				inode->i_size = PAGE_SIZE;
+				inode->i_fop = &configfs_file_operations;
+			}
 			break;
 		}
 	}
 
-	if (!found) {
-		/*
-		 * If it doesn't exist and it isn't a NOT_PINNED item,
-		 * it must be negative.
-		 */
-		d_add(dentry, NULL);
-		return NULL;
-	}
-
-	return ERR_PTR(err);
+	d_add(dentry, inode);
+	return NULL;
 }
 
 /*
-- 
2.30.2

Re: [PATCH 3/4] configfs: fold configfs_attach_attr into configfs_lookup

[Joel Becker]

Acked-by: Joel Becker <jlbec@evilplan.org>

On Wed, Aug 25, 2021 at 08:49:05AM +0200, Christoph Hellwig wrote:
> This makes it more clear what gets added to the dcache and prepares
> for an additional locking fix.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/configfs/dir.c | 73 ++++++++++++++++-------------------------------
>  1 file changed, 24 insertions(+), 49 deletions(-)
> 
> diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
> index 5d58569f0eea..fc20bd8a6337 100644
> --- a/fs/configfs/dir.c
> +++ b/fs/configfs/dir.c
> @@ -45,7 +45,7 @@ static void configfs_d_iput(struct dentry * dentry,
>  		/*
>  		 * Set sd->s_dentry to null only when this dentry is the one
>  		 * that is going to be killed.  Otherwise configfs_d_iput may
> -		 * run just after configfs_attach_attr and set sd->s_dentry to
> +		 * run just after configfs_lookup and set sd->s_dentry to
>  		 * NULL even it's still in use.
>  		 */
>  		if (sd->s_dentry == dentry)
> @@ -417,44 +417,13 @@ static void configfs_remove_dir(struct config_item * item)
>  	dput(dentry);
>  }
>  
> -
> -/* attaches attribute's configfs_dirent to the dentry corresponding to the
> - * attribute file
> - */
> -static int configfs_attach_attr(struct configfs_dirent * sd, struct dentry * dentry)
> -{
> -	struct configfs_attribute * attr = sd->s_element;
> -	struct inode *inode;
> -
> -	spin_lock(&configfs_dirent_lock);
> -	dentry->d_fsdata = configfs_get(sd);
> -	sd->s_dentry = dentry;
> -	spin_unlock(&configfs_dirent_lock);
> -
> -	inode = configfs_create(dentry, (attr->ca_mode & S_IALLUGO) | S_IFREG);
> -	if (IS_ERR(inode)) {
> -		configfs_put(sd);
> -		return PTR_ERR(inode);
> -	}
> -	if (sd->s_type & CONFIGFS_ITEM_BIN_ATTR) {
> -		inode->i_size = 0;
> -		inode->i_fop = &configfs_bin_file_operations;
> -	} else {
> -		inode->i_size = PAGE_SIZE;
> -		inode->i_fop = &configfs_file_operations;
> -	}
> -	d_add(dentry, inode);
> -	return 0;
> -}
> -
>  static struct dentry * configfs_lookup(struct inode *dir,
>  				       struct dentry *dentry,
>  				       unsigned int flags)
>  {
>  	struct configfs_dirent * parent_sd = dentry->d_parent->d_fsdata;
>  	struct configfs_dirent * sd;
> -	int found = 0;
> -	int err;
> +	struct inode *inode = NULL;
>  
>  	if (dentry->d_name.len > NAME_MAX)
>  		return ERR_PTR(-ENAMETOOLONG);
> @@ -471,28 +440,34 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  		return ERR_PTR(-ENOENT);
>  
>  	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
> -		if (sd->s_type & CONFIGFS_NOT_PINNED) {
> -			const unsigned char * name = configfs_get_name(sd);
> +		if ((sd->s_type & CONFIGFS_NOT_PINNED) &&
> +		    !strcmp(configfs_get_name(sd), dentry->d_name.name)) {
> +			struct configfs_attribute *attr = sd->s_element;
> +			umode_t mode = (attr->ca_mode & S_IALLUGO) | S_IFREG;
>  
> -			if (strcmp(name, dentry->d_name.name))
> -				continue;
> +			spin_lock(&configfs_dirent_lock);
> +			dentry->d_fsdata = configfs_get(sd);
> +			sd->s_dentry = dentry;
> +			spin_unlock(&configfs_dirent_lock);
>  
> -			found = 1;
> -			err = configfs_attach_attr(sd, dentry);
> +			inode = configfs_create(dentry, mode);
> +			if (IS_ERR(inode)) {
> +				configfs_put(sd);
> +				return ERR_CAST(inode);
> +			}
> +			if (sd->s_type & CONFIGFS_ITEM_BIN_ATTR) {
> +				inode->i_size = 0;
> +				inode->i_fop = &configfs_bin_file_operations;
> +			} else {
> +				inode->i_size = PAGE_SIZE;
> +				inode->i_fop = &configfs_file_operations;
> +			}
>  			break;
>  		}
>  	}
>  
> -	if (!found) {
> -		/*
> -		 * If it doesn't exist and it isn't a NOT_PINNED item,
> -		 * it must be negative.
> -		 */
> -		d_add(dentry, NULL);
> -		return NULL;
> -	}
> -
> -	return ERR_PTR(err);
> +	d_add(dentry, inode);
> +	return NULL;
>  }
>  
>  /*
> -- 
> 2.30.2
> 

-- 

[PATCH 4/4] configfs: fix a race in configfs_lookup()

[Christoph Hellwig]

From: Sishuai Gong <sishuai@purdue.edu>

When configfs_lookup() is executing list_for_each_entry(),
it is possible that configfs_dir_lseek() is calling list_del().
Some unfortunate interleavings of them can cause a kernel NULL
pointer dereference error

Thread 1                  Thread 2
//configfs_dir_lseek()    //configfs_lookup()
list_del(&cursor->s_sibling);
                         list_for_each_entry(sd, ...)

Fix this by grabbing configfs_dirent_lock in configfs_lookup()
while iterating ->s_children.

Signed-off-by: Sishuai Gong <sishuai@purdue.edu>
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/configfs/dir.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index fc20bd8a6337..1466b5d01cbb 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -439,13 +439,13 @@ static struct dentry * configfs_lookup(struct inode *dir,
 	if (!configfs_dirent_is_ready(parent_sd))
 		return ERR_PTR(-ENOENT);
 
+	spin_lock(&configfs_dirent_lock);
 	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
 		if ((sd->s_type & CONFIGFS_NOT_PINNED) &&
 		    !strcmp(configfs_get_name(sd), dentry->d_name.name)) {
 			struct configfs_attribute *attr = sd->s_element;
 			umode_t mode = (attr->ca_mode & S_IALLUGO) | S_IFREG;
 
-			spin_lock(&configfs_dirent_lock);
 			dentry->d_fsdata = configfs_get(sd);
 			sd->s_dentry = dentry;
 			spin_unlock(&configfs_dirent_lock);
@@ -462,10 +462,11 @@ static struct dentry * configfs_lookup(struct inode *dir,
 				inode->i_size = PAGE_SIZE;
 				inode->i_fop = &configfs_file_operations;
 			}
-			break;
+			goto done;
 		}
 	}
-
+	spin_unlock(&configfs_dirent_lock);
+done:
 	d_add(dentry, inode);
 	return NULL;
 }
-- 
2.30.2

Re: [PATCH 4/4] configfs: fix a race in configfs_lookup()

[Joel Becker]

Acked-by: Joel Becker <jlbec@evilplan.org>

On Wed, Aug 25, 2021 at 08:49:06AM +0200, Christoph Hellwig wrote:
> From: Sishuai Gong <sishuai@purdue.edu>
> 
> When configfs_lookup() is executing list_for_each_entry(),
> it is possible that configfs_dir_lseek() is calling list_del().
> Some unfortunate interleavings of them can cause a kernel NULL
> pointer dereference error
> 
> Thread 1                  Thread 2
> //configfs_dir_lseek()    //configfs_lookup()
> list_del(&cursor->s_sibling);
>                          list_for_each_entry(sd, ...)
> 
> Fix this by grabbing configfs_dirent_lock in configfs_lookup()
> while iterating ->s_children.
> 
> Signed-off-by: Sishuai Gong <sishuai@purdue.edu>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/configfs/dir.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
> index fc20bd8a6337..1466b5d01cbb 100644
> --- a/fs/configfs/dir.c
> +++ b/fs/configfs/dir.c
> @@ -439,13 +439,13 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  	if (!configfs_dirent_is_ready(parent_sd))
>  		return ERR_PTR(-ENOENT);
>  
> +	spin_lock(&configfs_dirent_lock);
>  	list_for_each_entry(sd, &parent_sd->s_children, s_sibling) {
>  		if ((sd->s_type & CONFIGFS_NOT_PINNED) &&
>  		    !strcmp(configfs_get_name(sd), dentry->d_name.name)) {
>  			struct configfs_attribute *attr = sd->s_element;
>  			umode_t mode = (attr->ca_mode & S_IALLUGO) | S_IFREG;
>  
> -			spin_lock(&configfs_dirent_lock);
>  			dentry->d_fsdata = configfs_get(sd);
>  			sd->s_dentry = dentry;
>  			spin_unlock(&configfs_dirent_lock);
> @@ -462,10 +462,11 @@ static struct dentry * configfs_lookup(struct inode *dir,
>  				inode->i_size = PAGE_SIZE;
>  				inode->i_fop = &configfs_file_operations;
>  			}
> -			break;
> +			goto done;
>  		}
>  	}
> -
> +	spin_unlock(&configfs_dirent_lock);
> +done:
>  	d_add(dentry, inode);
>  	return NULL;
>  }
> -- 
> 2.30.2
> 

--