--parallel=0 is a counterintuitive way of enabling parallelisation
Currently, to enable parallelisation in the general case, you have to pass --parallel=0
to g-d-t-r, which tells it to enable parallelisation up to the number of processor cores. Without passing --parallel
at all, the default is 1 processor core (i.e. no parallelisation).
However, passing --parallel=0
makes it look like you’re telling it to disable parallelisation.
How about enabling parallelisation by default, and also accepting --parallel
(with no value) as meaning “enable parallelisation to the number of processor cores”?
(I don’t plan on working on this.)