On 15 February 2017 at 13:34, Peter Maydell <peter.mayd...@linaro.org> wrote: > On 15 February 2017 at 12:46, Alex Bennée <alex.ben...@linaro.org> wrote: >> Can we not calculate a vector index rather than abusing the meaning of >> offset while switching on it? > > Yeah, we could. (This is just a case where I thought "I could > rewrite the code Michael did initially, but it doesn't quite > reach my threshold for doing that just because it's not the > way I'd have written it.".)
FWIW, it took me five attempts at rewriting these loops before I got a version that actually worked. I think I'm going to take that as justification for my bias towards not rewriting code that already works :-) (The change I ended up with just uses a local 'startvec' instead of modifying 'offset'.) thanks -- PMM
