fs: take i_mutex during prepare_binprm for set[ug]id executables

This prevents a race between chown() and execve(), where chowning a
setuid-user binary to root would momentarily make the binary setuid
root.

This patch was mostly written by Linus Torvalds.

Signed-off-by: Jann Horn <jann@thejh.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 48 insertions, 28 deletions

diff --git a/fs/exec.c b/fs/exec.c
index 02bfd980a40c..49a1c61433b7 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1275,6 +1275,53 @@ static void check_unsafe_exec(struct linux_binprm *bprm)
         spin_unlock(&p->fs->lock);
 }
 
+static void bprm_fill_uid(struct linux_binprm *bprm)
+{
+        struct inode *inode;
+        unsigned int mode;
+        kuid_t uid;
+        kgid_t gid;
+
+        /* clear any previous set[ug]id data from a previous binary */
+        bprm->cred->euid = current_euid();
+        bprm->cred->egid = current_egid();
+
+        if (bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID)
+                return;
+
+        if (task_no_new_privs(current))
+                return;
+
+        inode = file_inode(bprm->file);
+        mode = READ_ONCE(inode->i_mode);
+        if (!(mode & (S_ISUID|S_ISGID)))
+                return;
+
+        /* Be careful if suid/sgid is set */
+        mutex_lock(&inode->i_mutex);
+
+        /* reload atomically mode/uid/gid now that lock held */
+        mode = inode->i_mode;
+        uid = inode->i_uid;
+        gid = inode->i_gid;
+        mutex_unlock(&inode->i_mutex);
+
+        /* We ignore suid/sgid if there are no mappings for them in the ns */
+        if (!kuid_has_mapping(bprm->cred->user_ns, uid) ||
+                 !kgid_has_mapping(bprm->cred->user_ns, gid))
+                return;
+
+        if (mode & S_ISUID) {
+                bprm->per_clear |= PER_CLEAR_ON_SETID;
+                bprm->cred->euid = uid;
+        }
+
+        if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) {
+                bprm->per_clear |= PER_CLEAR_ON_SETID;
+                bprm->cred->egid = gid;
+        }
+}
+
 /*
  * Fill the binprm structure from the inode.
  * Check permissions, then read the first 128 (BINPRM_BUF_SIZE) bytes
@@ -1283,36 +1330,9 @@ static void check_unsafe_exec(struct linux_binprm *bprm)
  */
 int prepare_binprm(struct linux_binprm *bprm)
 {
-        struct inode *inode = file_inode(bprm->file);
-        umode_t mode = inode->i_mode;
         int retval;
 
-
-        /* clear any previous set[ug]id data from a previous binary */
-        bprm->cred->euid = current_euid();
-        bprm->cred->egid = current_egid();
-
-        if (!(bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID) &&
-            !task_no_new_privs(current) &&
-            kuid_has_mapping(bprm->cred->user_ns, inode->i_uid) &&
-            kgid_has_mapping(bprm->cred->user_ns, inode->i_gid)) {
-                /* Set-uid? */
-                if (mode & S_ISUID) {
-                        bprm->per_clear |= PER_CLEAR_ON_SETID;
-                        bprm->cred->euid = inode->i_uid;
-                }
-
-                /* Set-gid? */
-                /*
-                 * If setgid is set but no group execute bit then this
-                 * is a candidate for mandatory locking, not a setgid
-                 * executable.
-                 */
-                if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) {
-                        bprm->per_clear |= PER_CLEAR_ON_SETID;
-                        bprm->cred->egid = inode->i_gid;
-                }
-        }
+        bprm_fill_uid(bprm);
 
         /* fill in binprm security blob */
         retval = security_bprm_set_creds(bprm);