To: linux-kernel@vger.rutgers.edu Subject: revision control for the kernel (BitKeeper) From: lm@bitmover.com (Larry McVoy) Date: Sun, 21 Feb 1999 01:14:26 -0800 Most of you know we've been working on the next gen revision control for the kernel for a while now. For those of you who don't know, BitKeeper is an Open Source distributed revision control system which I claim is a substantial step forward from CVS. http://www.bitkeeper.com and http://www.bitmover.com/talks/linuxworld have more info. It's nearing a state where it could be put into use but there are some open issues. I'd like to discuss these issues, some of which are Linux specific, with people who are interested in this sort of thing. We could discuss it here, but I've set up an alias, kernel@bitmover.com, which is probably a better place. It's a majordomo alias. I've added some of the core kernel developers to that list, other people should add themselves. I'm especially interested in having some people with low bandwidth connections and disk space concerns be on the list. The bandwidth isn't a big problem, but the disk space is an issue. It does use more space to have this stuff under revision control and I want to discuss that to see if it is acceptable to the community. Here's an example: consider just the Historic kernels (0.1 .. 0.99.15, 38 in all). Size What 17M All 38 tarballs gzipped 11M All 38 versions checked into BitKeeper 1.8M All 38 versions checked into BitKeeper and bzip2-ed .5M The 0.99.15 tarball, gzipped So there is good news and not so good news. The good news is that if we are just talking tarballs, we can have all the history in less space than all the tarballs put together. And about 10x less space if we are talking about a tarball of the repository. The bad news is that I have a feeling that we might be looking at a 100-200MB repository for all the history going back to the beginning. We haven't converted all of it yet, I'll let you know the results as soon as we do. Anyway, I'd love to get your thoughts, please join the list and we'll figure it out. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/