Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762221AbYA3Sxh (ORCPT ); Wed, 30 Jan 2008 13:53:37 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756272AbYA3Sx1 (ORCPT ); Wed, 30 Jan 2008 13:53:27 -0500 Received: from rn-out-0910.google.com ([64.233.170.187]:29030 "EHLO rn-out-0910.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752744AbYA3Sx0 (ORCPT ); Wed, 30 Jan 2008 13:53:26 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth; b=cwMgJPvnfgBrD9BdxpQPJccatGtiDl786tgDKcBSOM71l67b33LH6zW5aLhMzcm2WrUpZStxtfcKMAqdk4AAo5vEYb5ySg93BYEtI7PKmrCGMzFyCx53oMiTmektDdYKXHo4X2MIlmuwyqD0wfHqhQz6i8eJLS2AHCtK1CdOCyY= Message-ID: Date: Wed, 30 Jan 2008 11:53:24 -0700 From: "Dan Williams" To: "Haavard Skinnemoen" Subject: Re: [RFC v2 4/5] dmaengine: Driver for the Synopsys DesignWare DMA controller Cc: linux-kernel@vger.kernel.org, "Shannon Nelson" , "David Brownell" , kernel@avr32linux.org, "Francis Moreau" , "Paul Mundt" , "Vladimir A. Barinov" , "Pierre Ossman" In-Reply-To: <1201630213-31900-5-git-send-email-hskinnemoen@atmel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <1201630213-31900-1-git-send-email-hskinnemoen@atmel.com> <1201630213-31900-2-git-send-email-hskinnemoen@atmel.com> <1201630213-31900-3-git-send-email-hskinnemoen@atmel.com> <1201630213-31900-4-git-send-email-hskinnemoen@atmel.com> <1201630213-31900-5-git-send-email-hskinnemoen@atmel.com> X-Google-Sender-Auth: 0215696e3d946f1f Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1068 Lines: 24 On Jan 29, 2008 11:10 AM, Haavard Skinnemoen wrote: [..] > The dmatest client shows no problems, but the performance is not as > good as it should be yet -- iperf shows a slight slowdown when > enabling TCP receive copy offload. This is probably because the > controller is set up to always do byte transfers; I'll try to optimize > this, but if someone can tell me if there any guaranteed alignment > requirements for the users of the DMA engine API, that would help a > lot. > dmaengine punts to the dma-mapping api. So no, there are no alignment guarantees. The performance loss is probably more related to the cache synchronization overkill of get_user_pages(). I/O incoherent architectures end up synchronizing entire pages when we only need to sync a kilobyte or two in this path. -- Dan -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/