From: Evgeniy Polyakov Subject: Re: [RFC] [PATCH v2 0/4] Parallel IPsec Date: Sat, 25 Apr 2009 14:10:19 +0400 Message-ID: <20090425101019.GA28498@ioremap.net> References: <20090424102451.GB12680@secunet.com> <20090425083849.GD26070@ioremap.net> <20090425092142.GA5830@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Steffen Klassert , David Miller , linux-crypto@vger.kernel.org To: Herbert Xu Return-path: Received: from cs-studio.ru ([195.178.208.66]:35905 "EHLO tservice.net.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751750AbZDYKK3 (ORCPT ); Sat, 25 Apr 2009 06:10:29 -0400 Content-Disposition: inline In-Reply-To: <20090425092142.GA5830@gondor.apana.org.au> Sender: linux-crypto-owner@vger.kernel.org List-ID: On Sat, Apr 25, 2009 at 05:21:42PM +0800, Herbert Xu (herbert@gondor.apana.org.au) wrote: > > Why can't it be used by default for all crypto operations instead of > > synchronous one? > > PCI-based drivers will not benefit from spreading the requests > across CPUs. If anything they will suffer from the synchronisation. What's the deal PCI drivers have with CPUs data comes from/to? They do not touch cache, just run DMA transfer and complete the request. -- Evgeniy Polyakov