The patch titled proc: fix races against execve() of /proc/PID/fd** has been added to the -mm tree. Its filename is proc-fix-races-against-execve-of-proc-pid-fd.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: proc: fix races against execve() of /proc/PID/fd** From: Vasiliy Kulikov <segoon@xxxxxxxxxxxx> fd* files are restricted to the task's owner, and other users may not get direct access to them. But one may open any of these files and run any setuid program, keeping opened file descriptors. As there are permission checks on open(), but not on readdir() and read(), operations on the kept file descriptors will not be checked. It makes it possible to violate procfs permission model. Reading fdinfo/* may disclosure current fds' position and flags, reading directory contents of fdinfo/ and fd/ may disclosure the number of opened files by the target task. This information is not sensible per se, but it can reveal some private information (like length of a password stored in a file) under certain conditions. Used existing (un)lock_trace functions to check for ptrace_may_access(), but instead of using EPERM return code from it use EACCES to be consistent with existing proc_pid_follow_link()/proc_pid_readlink() return code. If they differ, attacker can guess what fds exist by analyzing stat() return code. Patched handlers: stat() for fd/*, stat() and read() for fdindo/*, readdir() and lookup() for fd/ and fdinfo/. Signed-off-by: Vasiliy Kulikov <segoon@xxxxxxxxxxxx> Cc: <stable@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- fs/proc/base.c | 146 +++++++++++++++++++++++++++++++++-------------- 1 file changed, 103 insertions(+), 43 deletions(-) diff -puN fs/proc/base.c~proc-fix-races-against-execve-of-proc-pid-fd fs/proc/base.c --- a/fs/proc/base.c~proc-fix-races-against-execve-of-proc-pid-fd +++ a/fs/proc/base.c @@ -1665,12 +1665,46 @@ out: return error; } +static int proc_pid_fd_link_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct task_struct *task = get_proc_task(inode); + int rc; + + if (task == NULL) + return -ESRCH; + + rc = -EACCES; + if (lock_trace(task)) + goto out_task; + + generic_fillattr(inode, stat); + unlock_trace(task); + put_task_struct(task); + rc = 0; +out_task: + return rc; +} + static const struct inode_operations proc_pid_link_inode_operations = { .readlink = proc_pid_readlink, .follow_link = proc_pid_follow_link, .setattr = proc_setattr, }; +static const struct inode_operations proc_fdinfo_link_inode_operations = { + .setattr = proc_setattr, + .getattr = proc_pid_fd_link_getattr, +}; + +static const struct inode_operations proc_fd_link_inode_operations = { + .readlink = proc_pid_readlink, + .follow_link = proc_pid_follow_link, + .setattr = proc_setattr, + .getattr = proc_pid_fd_link_getattr, +}; + /* building an inode */ @@ -1902,49 +1936,61 @@ out: static int proc_fd_info(struct inode *inode, struct path *path, char *info) { - struct task_struct *task = get_proc_task(inode); - struct files_struct *files = NULL; + struct task_struct *task; + struct files_struct *files; struct file *file; int fd = proc_fd(inode); + int rc; - if (task) { - files = get_files_struct(task); - put_task_struct(task); - } - if (files) { - /* - * We are not taking a ref to the file structure, so we must - * hold ->file_lock. - */ - spin_lock(&files->file_lock); - file = fcheck_files(files, fd); - if (file) { - unsigned int f_flags; - struct fdtable *fdt; - - fdt = files_fdtable(files); - f_flags = file->f_flags & ~O_CLOEXEC; - if (FD_ISSET(fd, fdt->close_on_exec)) - f_flags |= O_CLOEXEC; - - if (path) { - *path = file->f_path; - path_get(&file->f_path); - } - if (info) - snprintf(info, PROC_FDINFO_MAX, - "pos:\t%lli\n" - "flags:\t0%o\n", - (long long) file->f_pos, - f_flags); - spin_unlock(&files->file_lock); - put_files_struct(files); - return 0; + task = get_proc_task(inode); + if (!task) + return -ENOENT; + + rc = -EACCES; + if (lock_trace(task)) + goto out_task; + + rc = -ENOENT; + files = get_files_struct(task); + if (files == NULL) + goto out_unlock; + + /* + * We are not taking a ref to the file structure, so we must + * hold ->file_lock. + */ + spin_lock(&files->file_lock); + file = fcheck_files(files, fd); + if (file) { + unsigned int f_flags; + struct fdtable *fdt; + + fdt = files_fdtable(files); + f_flags = file->f_flags & ~O_CLOEXEC; + if (FD_ISSET(fd, fdt->close_on_exec)) + f_flags |= O_CLOEXEC; + + if (path) { + *path = file->f_path; + path_get(&file->f_path); } - spin_unlock(&files->file_lock); - put_files_struct(files); - } - return -ENOENT; + if (info) + snprintf(info, PROC_FDINFO_MAX, + "pos:\t%lli\n" + "flags:\t0%o\n", + (long long) file->f_pos, + f_flags); + rc = 0; + } else + rc = -ENOENT; + spin_unlock(&files->file_lock); + put_files_struct(files); + +out_unlock: + unlock_trace(task); +out_task: + put_task_struct(task); + return rc; } static int proc_fd_link(struct inode *inode, struct path *path) @@ -2039,7 +2085,7 @@ static struct dentry *proc_fd_instantiat spin_unlock(&files->file_lock); put_files_struct(files); - inode->i_op = &proc_pid_link_inode_operations; + inode->i_op = &proc_fd_link_inode_operations; inode->i_size = 64; ei->op.proc_get_link = proc_fd_link; d_set_d_op(dentry, &tid_fd_dentry_operations); @@ -2071,7 +2117,12 @@ static struct dentry *proc_lookupfd_comm if (fd == ~0U) goto out; + result = ERR_PTR(-EACCES); + if (lock_trace(task)) + goto out; + result = instantiate(dir, dentry, task, &fd); + unlock_trace(task); out: put_task_struct(task); out_no_task: @@ -2091,23 +2142,28 @@ static int proc_readfd_common(struct fil retval = -ENOENT; if (!p) goto out_no_task; + + retval = -EACCES; + if (lock_trace(p)) + goto out; + retval = 0; fd = filp->f_pos; switch (fd) { case 0: if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) - goto out; + goto out_unlock; filp->f_pos++; case 1: ino = parent_ino(dentry); if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) - goto out; + goto out_unlock; filp->f_pos++; default: files = get_files_struct(p); if (!files) - goto out; + goto out_unlock; rcu_read_lock(); for (fd = filp->f_pos-2; fd < files_fdtable(files)->max_fds; @@ -2131,6 +2187,9 @@ static int proc_readfd_common(struct fil rcu_read_unlock(); put_files_struct(files); } + +out_unlock: + unlock_trace(p); out: put_task_struct(p); out_no_task: @@ -2208,6 +2267,7 @@ static struct dentry *proc_fdinfo_instan ei->fd = fd; inode->i_mode = S_IFREG | S_IRUSR; inode->i_fop = &proc_fdinfo_file_operations; + inode->i_op = &proc_fdinfo_link_inode_operations; d_set_d_op(dentry, &tid_fd_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ _ Patches currently in -mm which might be from segoon@xxxxxxxxxxxx are arch-arm-mach-ux500-mbox-db5500c-world-writable-sysfs-fifo-file.patch proc-fix-races-against-execve-of-proc-pid-fd.patch ipc-introduce-shm_rmid_forced-sysctl-testing.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html