From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Kok, Auke" Subject: Re: [E1000-devel] [PATCH 2/4] [RFC] e1000: make ioport free for supported devices Date: Thu, 06 Mar 2008 11:35:17 -0800 Message-ID: <47D04775.7020002@intel.com> References: <20080303223731.16344.94975.stgit@localhost.localdomain> <20080303223737.16344.16498.stgit@localhost.localdomain> <47CE84BB.1010408@garzik.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: e1000-devel@lists.sourceforge.net, netdev@vger.kernel.org To: Jeff Garzik Return-path: Received: from mga03.intel.com ([143.182.124.21]:7542 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754092AbYCFTjl (ORCPT ); Thu, 6 Mar 2008 14:39:41 -0500 In-Reply-To: <47CE84BB.1010408@garzik.org> Sender: netdev-owner@vger.kernel.org List-ID: Jeff Garzik wrote: > Auke Kok wrote: >> From: Tomohiro Kusumi >> >> Most e1000 devices do not need ioport access to function properly. >> This is desirable for a few massive homed systems where ioport >> resources become constrained. >> >> This patch is based on the code from Tomohiro Kusumi, heavily >> modified to fit PCI API changes and e1000 driver changes. >> >> Signed-off-by: Tomohiro Kusumi >> Signed-off-by: Auke Kok > > I don't think you need pci_select_bars()? It should just be > pci_enable_device_mem() in both paths AFAICS... honestly I don't know :) - I'll look into it. if that's the case then the patch will change significantly. thanks, Auke