"Robert P. J. Day" <rpjday@xxxxxxxxxxxxxx> writes: > > actually, a little experimentation suggests that this might be all i > need: > > - physically add <new file> > $ git add <new file> > $ git diff HEAD > patchfile > - physically remove <new file> > $ git rm <new file> > > what i get is a patch file that seems to fit the same pattern i'm used > to submitting. does the above look sane? > You are just looking at the changes in the index (from git add) versus the last known committed state (HEAD). It works, as long as you don't need any changelog or history associated with the patch. Really, it's the same as "diff -u /dev/null newfile", except it adds the git specific "new file mode" bits for git-apply. Eric -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
