From mboxrd@z Thu Jan 1 00:00:00 1970 From: Shmulik Hen Subject: Re: [Bonding-devel] Re: [bonding] compatibilty issues Date: Wed, 1 Oct 2003 11:49:04 +0300 Sender: netdev-bounce@oss.sgi.com Message-ID: <200310011149.04612.shmulik.hen@intel.com> References: Reply-To: shmulik.hen@intel.com Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit Cc: , Return-path: To: "David S. Miller" , "Chad N. Tindel" , , In-Reply-To: Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org On Wednesday 01 October 2003 10:05 am, David S. Miller wrote: > On Tue, 30 Sep 2003 17:36:50 -0400 > > "Chad N. Tindel" wrote: > > My recommendations are more towards the middle than either end. > > I would like to see us get rid of the _OLD ioctls in the 2.6 > > kernel specifically because it uses the SIOCDEVPRIVATE ioctls. > > ... > > > I would like to see them stay in 2.4 for the rest of the 2.4 tree > > specifically so that people who want to run on 3 year old systems > > can continue to do so without us breaking their world. > > I think this is fine, personally. > > I defer to Jeff for final judgment, he should be allowed to chime > in at least once more. > So here is what I did in the meantime: * Created a version for 2.4 that puts back all old compatibility stuff that was removed either during the propagation set or the cleanup set. * Created a version for 2.6 that puts back just the compatibility stuff that was removed in the propagation set (BOND_SETHWADDR, since we got a complaint from a RH9 user). * Removed the mention of the multicast param from the read-me. * Raised the ABI version to 2 so the new ifenslave keeps propagating IP settings to slaves for older drivers, and doesn't do that for new ones that contain Willy Tarreau's panic fix. As for not putting new stuff in 2.4.x kernels, here is where we stand; We believe new distributions based on 2.4.x kernels will keep showing for at least a year, probably longer, and that customers would like to see more bonding features in those distributions, so our intention is to keep getting new stuff into 2.4. We understand the drive to put new stuff into 2.6 and backport to 2.4 from time to time, but we'll really need to keep doing stuff the current way for a while. The cleanup stuff came up as a necessity before developing the next set of features, and those are all based on a cleaned-up bonding, so delaying the acceptance of the cleanup into 2.4 also delays our features acceptance. I'm waiting for the final word from everyone. I'll need to test the two new versions, but then I can release them accordingly. -- | Shmulik Hen Advanced Network Services | | Israel Design Center, Jerusalem | | LAN Access Division, Platform Networking | | Intel Communications Group, Intel corp. |