Currently the support for the reading δοχην ποιῃς includes 579. IGNTP Luke notes correctly that 579 reads ποιεις and that it is not alone in doing so (though it is alone in reading ποιεις in this word order). As the variant unit distinguishes between the two forms ποιησης and ποιης, it is probably better to include 579 in round brackets to avoid the conclusion that it supports the present subjunctive. I don't think there is a need to introduce the permutations with ποιεις in the apparatus though.

And the same is true for various witnesses given in support of ποιης δοχην.

Dirk,
This is a difficult situation--the point of the variation unit is to note the difference in word order, but there are indeed other instances in the apparatus where eta and epsilon differences are noted in subjunctives, e.g. Mk 13:7. This should be a situation for the editorial committee to discuss.  Thank you for your helpful comments--they are always welcomed!