From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756462Ab3AHObe (ORCPT ); Tue, 8 Jan 2013 09:31:34 -0500 Received: from tex.lwn.net ([70.33.254.29]:40583 "EHLO vena.lwn.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756424Ab3AHObc (ORCPT ); Tue, 8 Jan 2013 09:31:32 -0500 Date: Tue, 8 Jan 2013 07:31:30 -0700 From: Jonathan Corbet To: Marek Szyprowski Cc: Mauro Carvalho Chehab , Alessandro Rubini , federico.vaga@gmail.com, mchehab@infradead.org, pawel@osciak.com, hans.verkuil@cisco.com, giancarlo.asnaghi@st.com, linux-media@vger.kernel.org, linux-kernel@vger.kernel.org, s.nawrocki@samsung.com Subject: Re: [PATCH v3 2/4] videobuf2-dma-streaming: new videobuf2 memory allocator Message-ID: <20130108073130.38a8cc3d@lwn.net> In-Reply-To: <50EBC1C1.3060208@samsung.com> References: <3892735.vLSnhhCRFi@harkonnen> <1348484332-8106-1-git-send-email-federico.vaga@gmail.com> <1399400.izKZgEHXnP@harkonnen> <12929800.xFTBAueAE0@harkonnen> <20130106230947.GA17979@mail.gnudd.com> <20130107124050.3fc5031b@lwn.net> <20130107181500.24c56803@redhat.com> <50EBC1C1.3060208@samsung.com> Organization: LWN.net X-Mailer: Claws Mail 3.8.1 (GTK+ 2.24.10; x86_64-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 08 Jan 2013 07:50:41 +0100 Marek Szyprowski wrote: > > Couldn't this performance difference be due to the usage of GFP_DMA inside > > the VB2 code, like Federico's new patch series is proposing? > > > > If not, why are there a so large performance penalty? > > Nope, this was caused rather by a very poor CPU access to non-cached (aka > 'coherent') memory and the way the video data has been accessed/read > with CPU. Exactly. Uncached memory *hurts*, especially if you're having to touch it all with the CPU. jon