From: Tadeusz Struk Subject: Re: akcipher use Date: Sat, 27 Jun 2015 10:00:08 -0700 Message-ID: <558ED698.5080101@gmail.com> References: <1708764.x5PFhLSanv@tauon.atsec.com> <558C5444.4000105@intel.com> <1934055.of3DnBQRik@tachyon.chronox.de> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: linux-crypto@vger.kernel.org To: Stephan Mueller , Tadeusz Struk Return-path: Received: from mail-pd0-f172.google.com ([209.85.192.172]:34219 "EHLO mail-pd0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751073AbbF0RAY (ORCPT ); Sat, 27 Jun 2015 13:00:24 -0400 Received: by pdbep18 with SMTP id ep18so70486682pdb.1 for ; Sat, 27 Jun 2015 10:00:24 -0700 (PDT) In-Reply-To: <1934055.of3DnBQRik@tachyon.chronox.de> Sender: linux-crypto-owner@vger.kernel.org List-ID: Hi Stephan, On 06/27/2015 09:25 AM, Stephan Mueller wrote: > Ok, in this case, it would mean that rsa would point to the software > implementation and oaep(rsa) would point to the hardware. Would I be correct? All combinations are possible, and it is up to the implementation to decide. >> Yes, that's the main use case for this. We want to be able to accelerate SSL >> > handshakes. > Are you currently working on an AF_ALG interface? I'm currently working on adding rsa support to the qat driver. Once that will be finished my plan was to add AF_ALG interface for it. That will be in a few weeks time though since I'm on vacation next two weeks. If you have already started on AF_ALG you are more than welcome to do it. For me the most important thing will that it needs to be asynchronous. Regards, T