re: fs: Expand __fd_install_received() to accept fd

re: fs: Expand __fd_install_received() to accept fd

[Colin Ian King]

Hi,

static analysis with Coverity has detected a potential issue with the
following commit:

commit 8336af9ab8c5d64a309cbf72648054af61548899
Author: Kees Cook <keescook@chromium.org>
Date:   Wed Jun 10 08:46:58 2020 -0700

    fs: Expand __fd_install_received() to accept fd

Calling __fd_install_received() with fd >= 0 and ufd being non-null will
cause a put_user of an uninitialized new_fd hence potentially leaking
data on the stack back to the user.

static analysis is as follows:

1050 int __fd_install_received(int fd, struct file *file, int __user *ufd,
1051                          unsigned int o_flags)
1052 {
1053        struct socket *sock;

    1. var_decl: Declaring variable new_fd without initializer.

1054        int new_fd;
1055        int error;
1056
1057        error = security_file_receive(file);

    2. Condition error, taking false branch.

1058        if (error)
1059                return error;
1060

    3. Condition fd < 0, taking false branch.

1061        if (fd < 0) {
1062                new_fd = get_unused_fd_flags(o_flags);
1063                if (new_fd < 0)
1064                        return new_fd;
1065        }
1066

    4. Condition ufd, taking true branch.
1067        if (ufd) {

CID: Uninitialized scalar variable (UNINIT)5. uninit_use: Using
uninitialized value new_fd.

1068                error = put_user(new_fd, ufd);

Colin

Re: fs: Expand __fd_install_received() to accept fd

[Kees Cook]

On Mon, Jun 22, 2020 at 04:11:30PM +0100, Colin Ian King wrote:
> Hi,
> 
> static analysis with Coverity has detected a potential issue with the
> following commit:
> 
> commit 8336af9ab8c5d64a309cbf72648054af61548899
> Author: Kees Cook <keescook@chromium.org>
> Date:   Wed Jun 10 08:46:58 2020 -0700
> 
>     fs: Expand __fd_install_received() to accept fd
> 
> Calling __fd_install_received() with fd >= 0 and ufd being non-null will
> cause a put_user of an uninitialized new_fd hence potentially leaking
> data on the stack back to the user.
> 
> static analysis is as follows:
> 
> 1050 int __fd_install_received(int fd, struct file *file, int __user *ufd,
> 1051                          unsigned int o_flags)
> 1052 {
> 1053        struct socket *sock;
> 
>     1. var_decl: Declaring variable new_fd without initializer.
> 
> 1054        int new_fd;
> 1055        int error;
> 1056
> 1057        error = security_file_receive(file);
> 
>     2. Condition error, taking false branch.
> 
> 1058        if (error)
> 1059                return error;
> 1060
> 
>     3. Condition fd < 0, taking false branch.
> 
> 1061        if (fd < 0) {
> 1062                new_fd = get_unused_fd_flags(o_flags);
> 1063                if (new_fd < 0)
> 1064                        return new_fd;
> 1065        }
> 1066
> 
>     4. Condition ufd, taking true branch.
> 1067        if (ufd) {
> 
> CID: Uninitialized scalar variable (UNINIT)5. uninit_use: Using
> uninitialized value new_fd.
> 
> 1068                error = put_user(new_fd, ufd);
> 
> Colin

Eek. Thank you. Fixed with:


diff --git a/fs/file.c b/fs/file.c
index 9568bcfd1f44..c0bcf4c4fb12 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -963,12 +963,14 @@ int __fd_install_received(int fd, struct file *file, int __user *ufd,
 		new_fd = get_unused_fd_flags(o_flags);
 		if (new_fd < 0)
 			return new_fd;
-	}
+	} else
+		new_fd = fd;
 
 	if (ufd) {
 		error = put_user(new_fd, ufd);
 		if (error) {
-			put_unused_fd(new_fd);
+			if (fd < 0)
+				put_unused_fd(new_fd);
 			return error;
 		}
 	}
@@ -976,7 +978,6 @@ int __fd_install_received(int fd, struct file *file, int __user *ufd,
 	if (fd < 0)
 		fd_install(new_fd, get_file(file));
 	else {
-		new_fd = fd;
 		error = replace_fd(new_fd, file, o_flags);
 		if (error)
 			return error;

-- 
Kees Cook