[ODE] finite_rot_axis bugfix

[Russ Smith]

> Yes that should fix half the routine.

are you referring to your previous comment?:

> Apart from the redefinition though, irv could also be undefined at
> line 277 if code execution follows the else statement in line 256.

correct me if i'm wrong, but the statements at line 277 can not be
executed if code execution follows the else statement in line 256, as
both ifs check for the same thing.

russ.

-- 
Russell Smith
http://www.q12.org