From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH] add Attansic L2 PCI ID Date: Thu, 29 Mar 2007 14:56:38 -0400 Message-ID: <460C0BE6.8000700@garzik.org> References: <20070328234738.GA8042@devserv.devel.redhat.com> <460BAFB0.7000408@garzik.org> <460BC91A.4090407@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: xiong.huang@atheros.com, jacliburn@bellsouth.net, netdev@vger.kernel.org To: Chris Snook Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:55555 "EHLO mail.dvmed.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934324AbXC2S4o (ORCPT ); Thu, 29 Mar 2007 14:56:44 -0400 In-Reply-To: <460BC91A.4090407@redhat.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Chris Snook wrote: > Jeff Garzik wrote: >> Chris Snook wrote: >>> From: Chris Snook >>> >>> Add PCI ID for the Attansic L2 100 Mb ethernet adapter. >>> >>> Signed-off-by: Chris Snook >>> >>> --- linux-2.6.21-rc5.orig/include/linux/pci_ids.h 2007-03-27 >>> 23:26:50.000000000 -0400 >>> +++ linux-2.6.21-rc5/include/linux/pci_ids.h 2007-03-28 >>> 15:11:03.000000000 -0400 >>> @@ -2090,6 +2090,7 @@ >>> >>> #define PCI_VENDOR_ID_ATTANSIC 0x1969 >>> #define PCI_DEVICE_ID_ATTANSIC_L1 0x1048 >>> +#define PCI_DEVICE_ID_ATTANSIC_L2 0x2048 >> >> Actually you should be doing the reverse: >> >> Remove PCI_DEVICE_ID_ATTANSIC_L1, and replace the one place that uses >> it with the hexadecimal constant. >> >> Jeff > > We're working on integrating the driver for the L2 chip, so it will be > useful to symbolically distinguish between them. For now, adding the ID > serves to document the distinction between the L1 and L2 chips, as > they're alike enough that an atl1 driver hacked with the new PCI ID will > detect link status on the L2, even though it won't really work. By > getting the ID in now, we can distribute patches that don't touch core > code and won't need to be tweaked for submission. > > If pci_ids.h bloat is really a big deal, we can hold off until the L2 > patches are ready, but I don't see the harm in getting this out there now. It is current policy that we do not add PCI ID symbols that are only used in one place. It is pointless to add this. Jeff