From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Morton Subject: Re: [PATCH 2/2] forcedeth: Addition of new device id Date: Wed, 3 Jun 2009 14:52:38 -0700 Message-ID: <20090603145238.fd4ece19.akpm@linux-foundation.org> References: <4A257B1A.8080205@nvidia.com> <20090603132509.d3121eff.akpm@linux-foundation.org> <20090603.142647.149845259.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: aabdulla@nvidia.com, manfred@colorfullife.com, netdev@vger.kernel.org To: David Miller Return-path: Received: from smtp1.linux-foundation.org ([140.211.169.13]:56315 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753884AbZFCVwj (ORCPT ); Wed, 3 Jun 2009 17:52:39 -0400 In-Reply-To: <20090603.142647.149845259.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Wed, 03 Jun 2009 14:26:47 -0700 (PDT) David Miller wrote: > From: Andrew Morton > Date: Wed, 3 Jun 2009 13:25:09 -0700 > > > On Tue, 02 Jun 2009 15:18:50 -0400 > > Ayaz Abdulla wrote: > > > >> This patch adds support for new ethernet device in the MCP89 chipset. > >> > >> Signed-off-by: Ayaz Abdulla > >> > > > > It's best to combine these two patches into one, IMO. > > Andrew, we do not add new entries to pci_ids.h any more. Lots of people are still changing it. I believe that the approach we've been taking is: if the driver presently uses pci_ids.h and you're adding a new ID then go for it. But new drivers should use a private site for the IDs. And patches which switch old drivers from pci_ids over to a private site are approved of. > Just use the constant directly when using it in the driver > ID table. Well OK, but that means that forcedeth has most of its PCI IDs enumerated in pci_ids.h, but others hardwired in forcedeth.c. IOW, Ayaz owes us a "[patch] forcedeth: remove PCI IDs from pci_ids.h" ;)