From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753123Ab0CWChJ (ORCPT ); Mon, 22 Mar 2010 22:37:09 -0400 Received: from mail-yw0-f172.google.com ([209.85.211.172]:59571 "EHLO mail-yw0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752494Ab0CWChH convert rfc822-to-8bit (ORCPT ); Mon, 22 Mar 2010 22:37:07 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=DbchEk3DCOZSp70C4GCVJ12xxSQffZnLC7qIwQp9Drd9xBlBBBLj9E054SQSlRRy9y e6p6h1I7rr3GVPY+tTdeWt11SVPi8SY/e0pl9qDriHidhTJPxG4Y+IpHX743cNyQrQQz t6eNhLSCO/Z2tphN3sUeHRGZ3iG1wLMEiXPe4= MIME-Version: 1.0 In-Reply-To: <7b6bb4a51003221929i16673523nd5f558ccbaace397@mail.gmail.com> References: <20100320143240.GB2942@localhost.localdomain> <4BA7A3CF.8070503@trash.net> <628d1651003221834g543e3e6cl4ea39c0886cb4ba1@mail.gmail.com> <7b6bb4a51003221929i16673523nd5f558ccbaace397@mail.gmail.com> Date: Tue, 23 Mar 2010 10:37:06 +0800 Message-ID: <628d1651003221937q58ff39fel467636ab55531128@mail.gmail.com> Subject: Re: [PATCH] Netfilter: Fix integer overflow in net/ipv6/netfilter/ip6_tables.c From: wzt wzt To: Xiaotian Feng Cc: Patrick McHardy , linux-kernel@vger.kernel.org, netfilter-devel@vger.kernel.org Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > Patrick's point is that you're using "if (get.size >= INT_MAX / > sizeof(struct ipt_get_entries))" > So, did you find any chance that get.size * sizeof(struct > ipt_get_entries) >= INT_MAX ? > would you carefully read my explain??? get.size is copy from the user space, it can be set as 0x7fffffff, addition with sizeof(struct ipt_get_entries) can be overflow. > And, for the addition overflow, can it be caught by > > "if (*len != sizeof(struct ipt_get_entries) + get.size)" ??? > sizeof(struct ipt_get_entries) + get.size can be overflow as *len, get.size is control by user space with copy_from_user(). On Tue, Mar 23, 2010 at 10:29 AM, Xiaotian Feng wrote: > On Tue, Mar 23, 2010 at 9:34 AM, wzt wzt wrote: >>> I can see that the size might cause an overflow in the addition with >>> sizeof(struct ipt_get_entries) >> That's the integer overflow i pointed. >> get.size is copy from the user space,  it can be set as 0x7fffffff, >> addition with sizeof(struct ipt_get_entries) can be overflow. > > Patrick's point is that you're using "if (get.size >= INT_MAX / > sizeof(struct ipt_get_entries))" > So, did you find any chance that get.size * sizeof(struct > ipt_get_entries) >= INT_MAX ? > > And, for the addition overflow, can it be caught by > > "if (*len != sizeof(struct ipt_get_entries) + get.size)"  ??? > >> >>        if (*len != sizeof(struct ipt_get_entries) + get.size) { >>                duprintf("get_entries: %u != %zu\n", >>                         *len, sizeof(get) + get.size); >>                return -EINVAL; >>        } >> >> so, check get.size max value before addition with sizeof(struct >> ipt_get_entries) to prevent the integer overflow. >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at  http://vger.kernel.org/majordomo-info.html >> Please read the FAQ at  http://www.tux.org/lkml/ >> >