From: Andi Kleen Subject: Re: [dm-devel] [PATCH] DM-CRYPT: Scale to multiple CPUs Date: Mon, 31 May 2010 21:04:00 +0200 Message-ID: <20100531190400.GH10766@basil.fritz.box> References: <20100531160425.GA20344@basil.fritz.box> <4C03F04D.4040200@redhat.com> <20100531174225.GE10766@basil.fritz.box> <4C03FB8E.1060109@redhat.com> <20100531182724.GG10766@basil.fritz.box> <4C040636.8060702@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <4C040636.8060702@redhat.com> Sender: linux-kernel-owner@vger.kernel.org To: Milan Broz Cc: Andi Kleen , device-mapper development , herbert@gondor.hengli.com.au, linux-kernel@vger.kernel.org, agk@redhat.com, ak@linux.intel.com List-Id: dm-devel.ids > I mean how it is implemented now in crypto API, and I was almost > sure that aes-ni acceleration code uses cryptd (iow real asynchronous processing) > and also that not all CPU cores can run these instruction in parallel. I think you can configure it to use cryptd (or pcrypt), but it's not default and usually higher overhead. Each CPU core has its own support for AES-NI (it's part of the SSE units), there's no shared crypto hardware between cores. There are other systems with separate AES units, but they are considerably more obscure. -Andi -- ak@linux.intel.com -- Speaking for myself only.