diff options
author | Michael Brown <mcb30@etherboot.org> | 2005-03-24 13:54:46 +0000 |
---|---|---|
committer | Michael Brown <mcb30@etherboot.org> | 2005-03-24 13:54:46 +0000 |
commit | de5d935135bd536666b5fd03445b04d2ea82f1b4 (patch) | |
tree | d8dbe670b6d70ffe26bb312f39e63c78a81a7d71 /src | |
parent | afff2b50f1e887c6da2a391dff40285337b544a5 (diff) | |
download | ipxe-de5d935135bd536666b5fd03445b04d2ea82f1b4.tar.gz |
Merged this file into HEAD
Diffstat (limited to 'src')
-rw-r--r-- | src/README.cvs | 65 |
1 files changed, 65 insertions, 0 deletions
diff --git a/src/README.cvs b/src/README.cvs new file mode 100644 index 000000000..56a24f9fb --- /dev/null +++ b/src/README.cvs @@ -0,0 +1,65 @@ +Changes should be committed to the CVS HEAD only when they are in a +working state. The definition of "working" is somewhat liquid; a good +guiding principle is that anyone checking out HEAD should receive a +checkout of working software. + +When you want to work on changes that are likely to temporarily break +large swathes of code, you should probably work on a private branch. +Since CVS branching and merging is something of a black art, here are +some simple step-by-step instructions for creating and using a branch. + +To create your private branch: + + # Get most up-to-date tree before branching + cvs update + # Create a branch called "my-branch" + cvs tag -b my-branch + # Switch working copy to the "my-branch" branch + cvs update -r my-branch + +At this point you'll be on a branch called "my-branch". Any changes +you make will not affect people working on HEAD, or on other branches. + +Use name for your branch that is both descriptive and unique. +Starting the branch name with your SourceForge username +(e.g. "mcb30-realmode-redesign") is a good idea. + +When you want to merge the changes on your branch back into HEAD, do +the following: + + # Ensure there are no not-yet-checked-in modifications in your tree) + cvs -q update + # Tag the merge point in the "my-branch" branch + cvs tag -c my-branch-merge-1 + # Switch working copy back to HEAD + cvs update -A + # Merge changes from the branch + cvs update -j my-branch + # Commit merged changes to HEAD + cvs commit + +If you then want to continue working further on the "my-branch" branch, +do the following + + # Switch working copy back to the "my-branch" branch + cvs update -r my-branch + +and then when you want to merge some more changes back to HEAD: + + # Ensure there are no not-yet-checked-in modifications in your tree) + cvs -q update + # Tag the merge point in the "my-branch" branch + cvs tag -c my-branch-merge-2 + # Switch working copy back to HEAD + cvs update -A + # Merge changes from the branch + cvs update -j my-branch-merge-1 -j my-branch + # Commit merged changes to HEAD + cvs commit + +Note that the format of the "merge changes from the branch" command has +changed, because this time you need to only merge changes since the last +merge point. + +When you have finished with your branch and merged all the changes +back to HEAD, simply stop using the branch. |