From: Alexey Dobriyan Subject: Re: sha-512... Date: Wed, 15 Feb 2012 22:27:52 +0300 Message-ID: <20120215192751.GA27144@p183.telecom.by> References: <20120215040128.GA8738@gondor.apana.org.au> <20120215.001113.86424501006520715.davem@davemloft.net> <20120215051608.GA9126@gondor.apana.org.au> <20120215.002352.2053362114668441348.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: herbert@gondor.hengli.com.au, linux-crypto@vger.kernel.org, torvalds@linux-foundation.org To: David Miller Return-path: Received: from mail-bk0-f46.google.com ([209.85.214.46]:32862 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751610Ab2BOT16 (ORCPT ); Wed, 15 Feb 2012 14:27:58 -0500 Received: by bkcjm19 with SMTP id jm19so1292167bkc.19 for ; Wed, 15 Feb 2012 11:27:56 -0800 (PST) Content-Disposition: inline In-Reply-To: <20120215.002352.2053362114668441348.davem@davemloft.net> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Wed, Feb 15, 2012 at 12:23:52AM -0500, David Miller wrote: > From: Herbert Xu > Date: Wed, 15 Feb 2012 16:16:08 +1100 > > > OK, so we grew by 1136 - 888 = 248. Keep in mind that 128 of > > that is expected since we moved W onto the stack. > > Right. > > > I guess we could go back to the percpu solution, what do you > > think? > > I'm not entirely sure, we might have to. > > sha512 is notorious for generating terrible code with gcc on 32-bit > targets, so... The sha512 test in the glibc testsuite tends to > timeout on 32-bit sparc. :-) Cherrypicking ror64() commit largely fixes the issue (on sparc-defconfig): 00000000 : 0: 9d e3 bc 78 save %sp, -904, %sp git://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git b85a088f15f2070b7180735a231012843a5ac96c "crypto: sha512 - use standard ror64()"