Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 16 Jan 2002 06:55:27 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 16 Jan 2002 06:55:20 -0500 Received: from gnu.in-berlin.de ([192.109.42.4]:4868 "EHLO gnu.in-berlin.de") by vger.kernel.org with ESMTP id ; Wed, 16 Jan 2002 06:55:05 -0500 X-Envelope-From: kraxel@bytesex.org Date: Wed, 16 Jan 2002 10:57:47 +0100 From: Gerd Knorr To: Stephan von Krawczynski Cc: linux-kernel@vger.kernel.org Subject: Re: Memory problem with bttv driver Message-ID: <20020116105747.B1190@bytesex.org> In-Reply-To: <20020114210039.180c0438.skraw@ithnet.com> <20020115004205.A12407@werewolf.able.es> <20020115121424.10bb89b2.skraw@ithnet.com> <20020115142017.D8191@bytesex.org> <20020115161653.A9550@bytesex.org> <20020115173551.5a3fc051.skraw@ithnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20020115173551.5a3fc051.skraw@ithnet.com> User-Agent: Mutt/1.3.20i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > > > > Yes. Instead of remapping vmalloced kernel memory it gives you shared > > > > anonymous pages, then does zerocopy DMA using kiobufs. You may run in > > > > trouble with >4GB machines. > > > > > > Interesting. > > > What's the problem on > 4GB ? > > > > The bt878/848 is a 32bit PCI device, it simply can't go DMA to main > > memory above 4GB. At least on ia32, on architecures with a iommu > > (sparc, ...) it should work without trouble. > > Sorry, maybe I should have clarified: what is the problem with allocating those > pages below 4GB in a >4GB box? do_anonymous_page() may give you pages above 4GB. Need to write my own nopage handler to fix this. Gerd -- #define ENOCLUE 125 /* userland programmer induced race condition */ - 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/