We have to be careful when thinking of black holes "sucking", "vacuuming up", "gobbling", etc. Who was the sci-fi author (maybe there were several?) who used black holes as weapons? Some aliens dropped one on Earth (Or did we do it to ourselves? IDK) and it slowly, over decades, swallowed the matter of Earth. Its orbit was BELOW the surface of Earth, but being so massive, that didn't slow it down much. The question there would be: as it orbits the barycenter, how much matter flows into it? The viscosity of the matter (solid or molten) in the near-by (extreme) g field, along with its orbital velocity, would determine the inflow. Similarily, for a black hole inside a star, clearly the matter in its direct path would be (mostly) swallowed (I'd guess, I'd have to do the dynamics calculations to verify this, and I don't have the competence to do them) but near-by plasma would be limited to the regular laws of dynamics (in a general relativistic context). As an estimated lower bound, consider two disks, one the radius of the Sun, R, the other of radius of the BH, r. Then assume that each orbit removes the matter equivalent to πr². So the number of orbits required is R²/r². For a BH with an event horizon of 30 km (10X the Sun's mass, a large stellar sized BH) and given Sun's R=700,000 we have 7E5²/900 ~ 600,000 orbits. Assuming velocity is ~√(11GM/r) we can estimate that one orbit takes about 10 years at an orbital r = ½R, or it would take 6 million years for the Sun to be consumed. This is an extremely crude estimate, but it suggests that there's no reason that a bh couldn't exist temporarily inside a star. (Of course, the nasty (radiation) effects would be observable much sooner.)