From: Herbert Xu Subject: Re: [PATCH 2/3 v5] crypto: ux500: Add driver for HASH hardware Date: Wed, 18 Apr 2012 22:26:20 +0800 Message-ID: <20120418142620.GA23660@gondor.apana.org.au> References: <1334059426-6806-1-git-send-email-andreas.westin@stericsson.com> <1334059426-6806-3-git-send-email-andreas.westin@stericsson.com> <20120418074945.GA20681@gondor.apana.org.au> <4F8EC8A7.10300@stericsson.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "David S. Miller" , "linux-crypto@vger.kernel.org" , "kim.phillips@freescale.com" To: Andreas WESTIN Return-path: Received: from sting.hengli.com.au ([178.18.18.71]:51063 "EHLO fornost.hengli.com.au" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751274Ab2DRO01 (ORCPT ); Wed, 18 Apr 2012 10:26:27 -0400 Content-Disposition: inline In-Reply-To: <4F8EC8A7.10300@stericsson.com> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Wed, Apr 18, 2012 at 03:59:03PM +0200, Andreas WESTIN wrote: > > > On 2012-04-18 09:49, Herbert Xu wrote: > > > >Hmm, if this buffer is written to by the update function, then > >this doesn't work because the buffer resides in the tfm which > >can be used by multiple entities simultaneously. > > > >All writable data must reside in the request, not the tfm. > > I'm looking into it, I assume you mean "req->base.data" ? No, have a look at the helper ahash_request_ctx. You need to state how much memory you need via reqsize. Cheers, -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt