On Tue, May 28, 2013 at 1:02 PM, Luis R. Rodriguez <mcgrof@xxxxxxxxxxxxxxxx> wrote: > Right now we infer RELMOD_UPDATE based > on the output directory but inferring RELMOD_TYPE from the same target > output directory would be sloppy IMO so tackling these two with > --git-revision would need to be addressed. Addressing RELMOD_TYPE seems easy with --git-revision, add arguments additional arguments: -s -n -p -c -u as release mode specifications as arguments to be passed to gentree.py either as: a) standalone arguments, or; b) as a conglomeration of release modification types, ie something like --rel-mod=snpcu We just need a reasonable way to address RELMOD_UPDATE then. My current set of patches infer the backports tag that we'd use, if we rely on --git-revision we could just have an inferred map of what the respective backports tag should be, but that still does not address the RELMOD_UPDATE to use. If the backports tag is inferred from the linux tree used we have a few options: 1) Infer the RELMOD_UPDATE to use based on the latest backports tag, and verify it 2) Require the RELMOD_UPDATE to be specified it and then git verify it If we go down this road I like the first option. This however would likely require a map algorithm to be provided based on the linux tree to be used. This is fine for linux-next and linux-stable (given I merge Linus' tree on it but not sure if other folks are doing this) but lets consider that some folks might be using backports to generate releases (in theory) based on other trees like wireless-testing, let alone vendor trees (which I am not privy to) and not sure if that is a requirement. If the backports tag is *not* inferred from the linux tree to use we could ask --backports-revision I suppose, and from that infer the RELMOD_UPDATE. I'm inclined towards this option -- but then do we use git ls-tree and git rev-parse as you do with --git-revision or do we go with git reset and git paranoia? Do we have any use case for git paranoia then? Do we discard it? I'm still considering implementing this upstream as I find it useful to clean up my trees. Luis -- To unsubscribe from this list: send the line "unsubscribe backports" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
