[Libre-soc-dev] beginning adding subvl looping to ISACaller

[lkcl]

first cut did not cause "damage" i.e. once added as long as subvl=0 (vec1 i.e. no subvecs) everything remains fine.

i attempted a unit test with vec2 and it immediately brought up a "dumb" moment - SVSTATE has a subvl field *and* 24-bit RM of the prefix has a subvl field!

i feel so stoopid.

there is a need for 2 more bits in SVSTATE (a separate dest-subvl-loop-counter) so SVSTATE.subvl can be removed and replaced with SVSTATE.dsubstep.

l.