From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32151 invoked by alias); 25 Sep 2013 16:20:26 -0000 Mailing-List: contact libc-ports-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: libc-ports-owner@sourceware.org Received: (qmail 32137 invoked by uid 89); 25 Sep 2013 16:20:25 -0000 Received: from mail-pb0-f50.google.com (HELO mail-pb0-f50.google.com) (209.85.160.50) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-SHA encrypted) ESMTPS; Wed, 25 Sep 2013 16:20:25 +0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.7 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00,FREEMAIL_FROM autolearn=ham version=3.3.2 X-HELO: mail-pb0-f50.google.com Received: by mail-pb0-f50.google.com with SMTP id uo5so6278469pbc.9 for ; Wed, 25 Sep 2013 09:20:22 -0700 (PDT) X-Received: by 10.66.142.193 with SMTP id ry1mr111292pab.150.1380126021492; Wed, 25 Sep 2013 09:20:21 -0700 (PDT) Received: from [10.0.0.166] ([96.24.66.238]) by mx.google.com with ESMTPSA id py4sm48616242pbb.33.1969.12.31.16.00.00 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 25 Sep 2013 09:20:20 -0700 (PDT) References: <5242A79D.1030709@linaro.org> <52430AA4.70703@redhat.com> Mime-Version: 1.0 (1.0) In-Reply-To: <52430AA4.70703@redhat.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Message-Id: <34DAC322-9D37-411A-8C55-86B604F5799D@gmail.com> Cc: Will Newton , "libc-ports@sourceware.org" , "patches@linaro.org" From: pinskia@gmail.com Subject: Re: [PATCH] ARM: Add pointer guard support. Date: Wed, 25 Sep 2013 16:20:00 -0000 To: Carlos O'Donell X-IsSubscribed: yes X-SW-Source: 2013-09/txt/msg00133.txt.bz2 > On Sep 25, 2013, at 9:09 AM, "Carlos O'Donell" wrote: >=20 >> On 09/25/2013 05:06 AM, Will Newton wrote: >>=20 >> Add support for pointer mangling in glibc internal structures in C >> and assembler code. >>=20 >> Tested on armv7 with hard and soft thread pointers. >=20 > Have you measured the performance versus using the existing > global variable?=20 >=20 > TLS access on ARM is quite slow and it looks to me like it=20 > may be faster to use the global variable. Keep in mind that > the pointer guard and stack guard do not vary by thread. >=20 > 32-bit ARM is currently using a global variable e.g. > __pointer_chk_guard, all you need to do to make it work > is adjust the definitions of PTR_MANGLE and PTR_DEMANGLE > to reference the global symbol. >=20 > This is the second proposal for ARM (first was [1] for > AArch64) to support storing the a guard in the TCB, but > nobody has responded yet to my question about performance. T=20 I wonder the same question. Why move to using to tcb in the first place. = The only answer I can figure out is that is what x86 does but that is not a= lways the correct answer. >=20 > Cheers, > Carlos. >=20 > [1] https://sourceware.org/ml/libc-ports/2013-08/msg00052.html >=20