alt Hacker NewsMaybe, but also irrelevant to the discussion because whether you write mat[b * A + a] by hand or mat[b][a] and let the compiler frontend expand then makes no difference to the optimizer.
Maybe, but also irrelevant to the discussion because whether you write mat[b * A + a] by hand or mat[b][a] and let the compiler frontend expand then makes no difference to the optimizer.
You missed the point.
Those two representations are equivalent, yes. But that's not what flattening loops mean.