From: Joe Perches Subject: Re: [PATCH] crypto: sha1: modify sha1_update to use SHA1_BLOCK_SIZE Date: Wed, 25 May 2011 20:52:01 -0700 Message-ID: <1306381921.28839.17.camel@Joe-Laptop> References: <1306379477-24552-1-git-send-email-msb@chromium.org> <20110525.233411.1372538972166578285.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: msb@chromium.org, linux-kernel@vger.kernel.org, herbert@gondor.hengli.com.au, linux-crypto@vger.kernel.org To: David Miller Return-path: Received: from mail.perches.com ([173.55.12.10]:2057 "EHLO mail.perches.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755991Ab1EZDwD (ORCPT ); Wed, 25 May 2011 23:52:03 -0400 In-Reply-To: <20110525.233411.1372538972166578285.davem@davemloft.net> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Wed, 2011-05-25 at 23:34 -0400, David Miller wrote: > From: Mandeep Singh Baines > Date: Wed, 25 May 2011 20:11:17 -0700 > > Plus some other minor cleanup. > > Signed-off-by: Mandeep Singh Baines > The temp[] buffer is explicitly places inside the inner most > basic block so that the compiler doesn't allocate the stack > space unless that code path is taken. Does any version of gcc manage to do that? Regardless, it's still a good idea to keep declarations in the use scope.