From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755864Ab1G0VwT (ORCPT ); Wed, 27 Jul 2011 17:52:19 -0400 Received: from mga02.intel.com ([134.134.136.20]:16747 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755779Ab1G0Vtg (ORCPT ); Wed, 27 Jul 2011 17:49:36 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.67,279,1309762800"; d="scan'208";a="32826617" From: Andi Kleen References: <20110727247.325703029@firstfloor.org> In-Reply-To: <20110727247.325703029@firstfloor.org> To: segoon@openwall.com, torvalds@linux-foundation.org, ak@linux.intel.com, linux-kernel@vger.kernel.org, stable@kernel.org, tim.bird@am.sony.com Subject: [PATCH] [96/99] proc: restrict access to /proc/PID/io Message-Id: <20110727214936.0A8432403FF@tassilo.jf.intel.com> Date: Wed, 27 Jul 2011 14:49:36 -0700 (PDT) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 2.6.35-longterm review patch. If anyone has any objections, please let me know. ------------------ From: Vasiliy Kulikov [ upstream commit 1d1221f375c94ef961ba8574ac4f85c8870ddd51 ] /proc/PID/io may be used for gathering private information. E.g. for openssh and vsftpd daemons wchars/rchars may be used to learn the precise password length. Restrict it to processes being able to ptrace the target process. ptrace_may_access() is needed to prevent keeping open file descriptor of "io" file, executing setuid binary and gathering io information of the setuid'ed process. Said to be CVE-2011-2495 Signed-off-by: Vasiliy Kulikov Signed-off-by: Linus Torvalds Signed-off-by: Andi Kleen Index: linux-2.6.35.y/fs/proc/base.c =================================================================== --- linux-2.6.35.y.orig/fs/proc/base.c +++ linux-2.6.35.y/fs/proc/base.c @@ -2514,6 +2514,9 @@ static int do_io_accounting(struct task_ struct task_io_accounting acct = task->ioac; unsigned long flags; + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + return -EACCES; + if (whole && lock_task_sighand(task, &flags)) { struct task_struct *t = task; @@ -2636,7 +2639,7 @@ static const struct pid_entry tgid_base_ REG("coredump_filter", S_IRUGO|S_IWUSR, proc_coredump_filter_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING - INF("io", S_IRUGO, proc_tgid_io_accounting), + INF("io", S_IRUSR, proc_tgid_io_accounting), #endif }; @@ -2972,7 +2975,7 @@ static const struct pid_entry tid_base_s REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING - INF("io", S_IRUGO, proc_tid_io_accounting), + INF("io", S_IRUSR, proc_tid_io_accounting), #endif };