From mboxrd@z Thu Jan 1 00:00:00 1970 From: Albert ARIBAUD Date: Fri, 09 Dec 2011 18:53:36 +0100 Subject: [U-Boot] Pull request: u-boot-tegra/master In-Reply-To: <4EE24751.9020900@nvidia.com> References: <7E91C59574E9954FA075F8D8CCDF78DB392DCD5D59@HQMAIL04.nvidia.com> <4EE065AE.4010007@aribaud.net> <4EE0E99C.1000404@nvidia.com> <4EE22B01.3070409@aribaud.net> <4EE24113.7020005@nvidia.com> <4EE243E6.1060501@aribaud.net> <4EE24751.9020900@nvidia.com> Message-ID: <4EE24B20.7070703@aribaud.net> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Le 09/12/2011 18:37, Stephen Warren a ?crit : >> Er... Rebase *does* 3-way merges when necessary and possible. I just saw >> it done while pulling tegra, samsung and imx requests. > > I don't think that's possible; with rebase, there is no common ancestor, > so you can't do a 3-way merge. The rebase conflicts end up being marked > up in the conflicting files in a similar fashion to a 3-way merge > though, but I think it's somewhat fake and just showing the differences > between the base that was in the patch being applied and current code, > which isn't quite the same thing. Actually there is a common ancestor in a rebase just as in a merge -- the common ancestor is an attribute of (or more exactly a computation based on) a set of commits, not of an operation. You'll find common ancestors in a rebase operation mentioned in , for instance. Amicalement, -- Albert.