Log in

No account? Create an account


« previous entry | next entry »
Mar. 25th, 2007 | 11:59 pm

The mozilla tree still confuses me. I really want to get at least some chunk of mozilla into a distributed version control system, so I can hack on a patch or two and keep it current with the CVS HEAD.

Why do I want complicated things to work? This has been quite frustrating. Especially since I now suspect I just wasted a large chunk of time because my various types of mozilla checkouts weren't building because of my lingering build directory.

Note to self: If mozilla is having trouble building perhaps a rm -rf $MOZ_OBJDIR might improve things.

Another note, git's branching capabilities look really promising for making it easy to be working on a couple of different combinations of mozilla products and patches and accelerate my build with ccache. Robert O'Callahan had some hints in Sharing binaries across multiple Mozilla trees with ccache.

$ git checkout proxy-patch master
# (wait a bit for the delta between proxy and master to show up)
$ make -f client.mk build

seems far more likely to make ccache think a particular source-file is the same. (And since a fresh build of mozilla/browser on one core of my macbook can take a number of hours, caching build products would be a big win.

Link | Leave a comment |

Comments {0}