The Planck dust measurement in pretty damning, but it is not the final word.
(1) Planck measured the dust contamination with greatest sensitivity at 353 GHz. It was not sensitive enough to measure the dust signal at 150 GHz, where BICEP was observing. They had to extrapolate the dust contribution from the higher frequency to the lower. This is actually a pretty big extrapolation, since the dust emission at 150 GHz is less than 1% of the dust emission at 353 GHz.
(2) The uncertainty in the dust emission amplitude is still pretty high, so the Planck measurement is consistent with an "all dust" model, or with a "mostly dust" model, or with a "mostly primordial, with some dust" model. It does pretty conclusively rule out a "no dust" model.
(3) They have not released the results of a joint analysis of Planck and BICEP2, which is what is necessary to actually shed some light on exactly how much of the BICEP2 signal is likely to be dust.
But it's clear that the BICEP team was being over-optimistic in their assumptions about galactic dust, which is a bummer.