From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-0.8 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_PASS, URIBL_BLOCKED autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E90FDC43144 for ; Tue, 26 Jun 2018 21:19:49 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 5BE4926F68 for ; Tue, 26 Jun 2018 21:19:49 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=efficios.com header.i=@efficios.com header.b="fx+zf32G" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 5BE4926F68 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=efficios.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752547AbeFZVTr (ORCPT ); Tue, 26 Jun 2018 17:19:47 -0400 Received: from mail.efficios.com ([167.114.142.138]:41196 "EHLO mail.efficios.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752048AbeFZVTp (ORCPT ); Tue, 26 Jun 2018 17:19:45 -0400 Received: from localhost (ip6-localhost [IPv6:::1]) by mail.efficios.com (Postfix) with ESMTP id 4ACF422DBCD; Tue, 26 Jun 2018 17:19:45 -0400 (EDT) Received: from mail.efficios.com ([IPv6:::1]) by localhost (mail02.efficios.com [IPv6:::1]) (amavisd-new, port 10032) with ESMTP id 6XbLYpqVVprG; Tue, 26 Jun 2018 17:19:44 -0400 (EDT) Received: from localhost (ip6-localhost [IPv6:::1]) by mail.efficios.com (Postfix) with ESMTP id B9AF722DBC5; Tue, 26 Jun 2018 17:19:44 -0400 (EDT) DKIM-Filter: OpenDKIM Filter v2.10.3 mail.efficios.com B9AF722DBC5 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=efficios.com; s=default; t=1530047984; bh=3Osi75krxmrwfOBUSAFYSuCJ6gcVEwcL5JBOKaXIm4s=; h=Date:From:To:Message-ID:MIME-Version; b=fx+zf32Gr6P7oIelhA6ErF2nRUUrtwI2tMdIAOLjOiUEbHxhsCiz0I7Gs4kb5dBCJ AZmhjVn5j/iqSPIyf9/q37wNsRDwGMAsa79+4zfGqHeVba4BTqBeWdEDkRdpwPz0JJ HMUTY1EY7dNvXJoAYkEePOfl+24l+phYYZR2G3+uoOHcydR3XDw5UB0vXmLd7A5Z/T VMEd+TOualuDm3GnVfiaAIVoftoOFSIvzSsIIaWtsFTCJ4cGPzRVQbyfdIvsFQTsOZ C3zYPM1nx1g5d9xXcjOJGtIbc892PTyAEyYZnE6oeA04Eoq7l/mCCsbvt2Bbmy1i2j qnOQsfYv5B5QA== X-Virus-Scanned: amavisd-new at efficios.com Received: from mail.efficios.com ([IPv6:::1]) by localhost (mail02.efficios.com [IPv6:::1]) (amavisd-new, port 10026) with ESMTP id u6ORow84uDEK; Tue, 26 Jun 2018 17:19:44 -0400 (EDT) Received: from mail02.efficios.com (mail02.efficios.com [167.114.142.138]) by mail.efficios.com (Postfix) with ESMTP id A448322DBBE; Tue, 26 Jun 2018 17:19:44 -0400 (EDT) Date: Tue, 26 Jun 2018 17:19:44 -0400 (EDT) From: Mathieu Desnoyers To: Andy Lutomirski Cc: linux-arch , Peter Zijlstra , Boqun Feng , linux-kernel , "Paul E. McKenney" , Thomas Gleixner Message-ID: <1453026995.6388.1530047984544.JavaMail.zimbra@efficios.com> In-Reply-To: References: <1514459655.4190.1530034687884.JavaMail.zimbra@efficios.com> <170076903.5015.1530038711536.JavaMail.zimbra@efficios.com> <1277536320.5963.1530042608296.JavaMail.zimbra@efficios.com> <1352608225.6039.1530043932895.JavaMail.zimbra@efficios.com> Subject: Re: rseq: How to test for compat task at signal delivery MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [167.114.142.138] X-Mailer: Zimbra 8.8.8_GA_2096 (ZimbraWebClient - FF52 (Linux)/8.8.8_GA_1703) Thread-Topic: rseq: How to test for compat task at signal delivery Thread-Index: J9wU3OUAVFQPk6dzYLDpUzy43Ov/XQ== Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org ----- On Jun 26, 2018, at 4:46 PM, Andy Lutomirski luto@amacapital.net wrote: > On Tue, Jun 26, 2018 at 1:12 PM Mathieu Desnoyers > wrote: >> >> ----- On Jun 26, 2018, at 3:55 PM, Andy Lutomirski luto@amacapital.net wrote: >> >> > On Tue, Jun 26, 2018 at 12:50 PM Mathieu Desnoyers >> > wrote: >> >> >> >> ----- On Jun 26, 2018, at 3:32 PM, Andy Lutomirski luto@amacapital.net wrote: >> >> >> >> > On Tue, Jun 26, 2018 at 11:45 AM Mathieu Desnoyers >> >> > wrote: >> >> >> >> >> >> ----- On Jun 26, 2018, at 1:38 PM, Mathieu Desnoyers >> >> >> mathieu.desnoyers@efficios.com wrote: >> >> >> >> >> >> > Hi Andy, >> >> >> > >> >> >> > I would like to make the behavior rseq on compat tasks more robust >> >> >> > by ensuring that kernel/rseq.c:rseq_get_rseq_cs() clears the high >> >> >> > bits of rseq_cs->abort_ip, rseq_cs->start_ip and >> >> >> > rseq_cs->post_commit_offset when a 32-bit binary is run on a 64-bit >> >> >> > kernel. >> >> >> > >> >> >> > The intent here is that if user-space has garbage rather than zeroes >> >> >> > in its struct rseq_cs fields padding, the behavior will be the same >> >> >> > whether the binary is run on 32-bit or 64 kernels. >> >> >> > >> >> >> > I know that internally, the kernel is making a transition from >> >> >> > is_compat_task() to in_compat_syscall(). >> >> >> > >> >> >> > I'm fine with using in_compat_syscall() when rseq_get_rseq_cs() is >> >> >> > invoked from a system call, but is it OK to call it when it is >> >> >> > invoked from signal delivery ? AFAIU, signals can be delivered >> >> >> > upon return from interrupt as well. >> >> >> > >> >> >> > If not, what strategy do you recommend for arch-agnostic code ? >> >> >> >> >> >> I think what we're missing here is a new "is_compat_frame(struct ksignal *ksig)" >> >> >> which I could use in the rseq code. I'll prepare a patch and we can discuss >> >> >> from there. >> >> >> >> >> > >> >> > That sounds about right. >> >> > >> >> > I'm confused, though. Wouldn't it be more consistent to just segfault >> >> > if the high 32 bits are not clear when rseq transitions to a 32-bit >> >> > context? If there's garbage in 64-bit mode, the program will crash. >> >> > Why should 32-bit mode be any different? >> >> >> >> Currently, if a 32-bit binary puts garbage in the high bits of >> >> start_ip, post_commit_offset, and abort_ip in >> >> >> >> include/uapi/linux/rseq.h: >> >> >> >> struct rseq_cs { >> >> /* Version of this structure. */ >> >> __u32 version; >> >> /* enum rseq_cs_flags */ >> >> __u32 flags; >> >> LINUX_FIELD_u32_u64(start_ip); >> >> /* Offset from start_ip. */ >> >> LINUX_FIELD_u32_u64(post_commit_offset); >> >> LINUX_FIELD_u32_u64(abort_ip); >> >> } __attribute__((aligned(4 * sizeof(__u64)))); >> > >> > This ABI isn't real ABI until a stable kernel happens, right? So how >> > about just making all those fields be u64? >> >> Good point. Unlike the rseq_cs field in the struct rseq TLS, those >> fields don't need to be word-sized/word-aligned, so we could simply >> declare them as __u64. >> >> > >> >> >> >> A 32-bit kernel just never reads the padding, thus in reality acting >> >> as if those were zeroes. However, a 64-bit kernel dealing with this >> >> 32-bit compat task will read that padding, handling those as very >> >> large values. >> > >> > Sounds like a design error. Have all kernels read the fields no >> > matter what. A 32-bit kernel will send SIGSEGV if the high bits are >> > set. A 64-bit kernel running compat userspace should make sure that a >> > 32-bit task dies if the high bits are set. >> >> If we end up declaring those as __u64, that approach makes sense. >> >> > >> >> >> >> We need to improve that by introducing a consistent behavior across >> >> native 32-bit kernels and 32-bit compat mode on 64-bit kernels. >> >> >> >> There are two ways to achieve this: either the 32-bit kernel validates >> >> the padding by killing the process if padding is non-zero, or the >> >> 64-bit kernel treats compat mode by zeroing the high bits of padding. >> >> >> >> If we look at system call interfaces in general, I think the usual >> >> approach is to clear the top bits whenever a value read from a >> >> compat task ends up being used as a pointer. This is why I am tempted >> >> to go for the "clear high bits" approach rather than killing the task. >> > >> > I think the modern preference is to use fields of fixed size rather >> > than long when UABI is involved. >> > >> > In any event, I think the test you want is user_64bit_mode(). >> >> Currently, user_64bit_mode is only implemented on x86. >> >> Should we introduce an architecture-agnostic user_64bit_mode(struct pt_regs *) >> which maps to is_compat_task() for non-x86 ? I'm just worried that ptrace >> code could try to use it from the context of another task and get mixed up. > > I'm not sure other archs can do this. It might need to have a > task_struct pointer, too. > > But I think the only actual consideration is that a lot of > architectures might fail to kill the task if the task is 32-bit and > regs->ip or regs->sp ends up with garbage in the high bits. Certainly > x86 is not consistent about this. So maybe a helper to fully validate > all 64 bits of ip and sp or perhaps helpers to set them and check for > full validity would be better. Like: > > void set_task_64bit_ip_or_signal(struct task_struct *, u64 value); > > that promises to actually signal the task if value is garbage? > > Let's ask linux-arch here. I'm not nearly familiar enough with the > nasty details of other compat-capable architectures. x86 is very, > very, very inconsistent about how what the high bits of the registers > mean, and there are cases where the "high bits" involved are actually > the high 48 bits, not the high 32 bits. Sigh. For the records, I just sent out 2 patches as RFC implementing the is_compat_frame() approach, and clearing the high bits for compat tasks. It has the merit to be straightforward and introduce few changes at this stage of the -rc. Thanks, Mathieu -- Mathieu Desnoyers EfficiOS Inc. http://www.efficios.com