Is now a good time to branch?

[Nick Roberts]

> The new gdb/gdba merge brought to my attention the fact that gdba seems not
> to like it when the .gdbinit file does funny things like set breakpoints
> (as is done in emacs/src/.gdbinit).

> It seems that it does not completely confuse gdba in the end, but there's
> at least some complaining going on as well as some aesthetically
> displeasing side effects.

I can't reproduce this consistently, but I think this mode sometimes get
confused by the word wrapping that GDB does to its output. In the short term
putting "set width 0" at the start of .gdbinit might solve it.  In the longer
term migrating more of the functionality to GDB/MI will eliminate such
problems as there is no wrapping with this (machine) interface. However as I
start to do this I am concerned about breaking existing functionality. So...

Would it be a good idea to create a branch from HEAD in the near future, say
after the image support for Carbon Emacs has been committed, so that new
features can be developed more freely?

Nick