From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757200AbYEGQu0 (ORCPT ); Wed, 7 May 2008 12:50:26 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752028AbYEGQt7 (ORCPT ); Wed, 7 May 2008 12:49:59 -0400 Received: from mail.issp.bas.bg ([195.96.236.10]:45727 "EHLO mail.issp.bas.bg" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751747AbYEGQt5 (ORCPT ); Wed, 7 May 2008 12:49:57 -0400 From: Marin Mitov Organization: Institute of Solid State Physics To: linux-kernel@vger.kernel.org Subject: CONFIG_HIGHMEM or CONFIG_HIGHMEM64G in illegal_highdma() Date: Wed, 7 May 2008 19:49:54 +0300 User-Agent: KMail/1.9.7 Cc: netdev@vger.kernel.org MIME-Version: 1.0 Content-Disposition: inline Message-Id: <200805071949.54583.mitov@issp.bas.bg> Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi all, The usage of NETIF_F_HIGHDMA in drivers/net/* make me think its purpose is to declare drivers's capability to address RAM above 4GB, OK? For x86_64 CONFIG_HIGHMEM is not defined, so the function illegal_highdma() found in net/core/dev.c returns 0 because x86_64 has IOMMU mapping all RAM. For x86_32 CONFIG_HIGHMEM is defined as (CONFIG_HIGHMEM4G || CONFIG_HIGHMEM64G) While CONFIG_HIGHMEM4G insures RAM is bellow 4GB some checks are unnecessarily done. Should we have CONFIG_HIGHMEM64G in illegal_highdma() instead of CONFIG_HIGHMEM? If so should I post a patch? Regards. Marin Mitov P.S. This is a resend, due to recent problems with http://lkml.org/