Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265470AbUFZE7I (ORCPT ); Sat, 26 Jun 2004 00:59:08 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S266365AbUFZE7I (ORCPT ); Sat, 26 Jun 2004 00:59:08 -0400 Received: from palrel13.hp.com ([156.153.255.238]:17866 "EHLO palrel13.hp.com") by vger.kernel.org with ESMTP id S265470AbUFZE7F (ORCPT ); Sat, 26 Jun 2004 00:59:05 -0400 From: David Mosberger MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16605.653.502075.164097@napali.hpl.hp.com> Date: Fri, 25 Jun 2004 21:58:53 -0700 To: Andi Kleen Cc: Terence Ripperda , discuss@x86-64.org, tiwai@suse.de, linux-kernel@vger.kernel.org, andrea@suse.de Subject: Re: 32-bit dma allocations on 64-bit platforms In-Reply-To: <20040624185156.GA19559@colin2.muc.de> References: <20040623234644.GC38425@colin2.muc.de> <20040624154429.GC8014@hygelac> <20040624185156.GA19559@colin2.muc.de> X-Mailer: VM 7.18 under Emacs 21.3.1 Reply-To: davidm@hpl.hp.com X-URL: http://www.hpl.hp.com/personal/David_Mosberger/ Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 836 Lines: 22 >>>>> On Thu, 24 Jun 2004 20:51:56 +0200, Andi Kleen said: Andi> A better IO_TLB_SHIFT would be 16 or 17. Careful. I see code like this: stride = (1 << (PAGE_SHIFT - IO_TLB_SHIFT)); You probably don't want IO_TLB_SHIFT > PAGE_SHIFT... Increasing io_tlb_nslabs should be no problem though (subject to memory availability). It can already by set via the "swiotlb" option. I doubt swiotlb is the right thing here, though, given the bw-demands of graphics. Too bad Nvidia cards don't support > 32 bit addressability and Intel chipsets don't support I/O MMUs... --david - 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/