From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751768AbdL1HCg (ORCPT ); Thu, 28 Dec 2017 02:02:36 -0500 Received: from [128.1.224.119] ([128.1.224.119]:48978 "EHLO ringil.hmeau.com" rhost-flags-FAIL-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1750981AbdL1HCf (ORCPT ); Thu, 28 Dec 2017 02:02:35 -0500 Date: Thu, 28 Dec 2017 18:02:22 +1100 From: Herbert Xu To: Eric Biggers Cc: linux-crypto@vger.kernel.org, "David S . Miller" , Josh Poimboeuf , Jussi Kivilinna , x86@kernel.org, linux-kernel@vger.kernel.org, syzkaller-bugs@googlegroups.com, Eric Biggers Subject: Re: [PATCH] crypto: x86/twofish-3way - Fix %rbp usage Message-ID: <20171228070222.GB9639@gondor.apana.org.au> References: <001a113f2cd26f3532055f0f4a79@google.com> <20171219004026.170565-1-ebiggers3@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20171219004026.170565-1-ebiggers3@gmail.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Dec 18, 2017 at 04:40:26PM -0800, Eric Biggers wrote: > From: Eric Biggers > > Using %rbp as a temporary register breaks frame pointer convention and > breaks stack traces when unwinding from an interrupt in the crypto code. > > In twofish-3way, we can't simply replace %rbp with another register > because there are none available. Instead, we use the stack to hold the > values that %rbp, %r11, and %r12 were holding previously. Each of these > values represents the half of the output from the previous Feistel round > that is being passed on unchanged to the following round. They are only > used once per round, when they are exchanged with %rax, %rbx, and %rcx. > > As a result, we free up 3 registers (one per block) and can reassign > them so that %rbp is not used, and additionally %r14 and %r15 are not > used so they do not need to be saved/restored. > > There may be a small overhead caused by replacing 'xchg REG, REG' with > the needed sequence 'mov MEM, REG; mov REG, MEM; mov REG, REG' once per > round. But, counterintuitively, when I tested "ctr-twofish-3way" on a > Haswell processor, the new version was actually about 2% faster. > (Perhaps 'xchg' is not as well optimized as plain moves.) > > Reported-by: syzbot > Signed-off-by: Eric Biggers Patch applied. Thanks. -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt