Well, I know there has been a lot of movement lately in the theory of the {1.d4 Nf6 2.c4 c5 3.d5 e6 4.Nc3 exd5 5.cxd5 d6 6.e4 g6 7.f4 Bg7 8.Bb5†} 8...Nbd7!? line, but I cannot bring myself to trust it. I think if you have a good understanding of typical Benoni themes you are likely to do well in the 8...Nfd7(!) lines, even if they might objectively be += (which is certainly not beyond dispute).
A lot of White players play the Flick-Knife because of its reputation, bash out what little of the theory they know then wonder why it is that Black's position isn't falling apart like the well-known Kasparov-Nunn game (which is available in the Nimzo & Benoni section).
The thing is, it is actually quite hard to handle the White side of this in my view. For the Black side, you can do no better than look at Watson's coverage in
The Gambit Guide to the Modern Benoni.
He recommends playing the lines with 9.a4 Qh4†, which I believe is the best approach for a number of reasons:
(1) Most importantly, I think it is objectively best, as inducing g2-g3 [White has some interesting ideas with Kf1!? instead but they are rare and Black should be able to handle them too] serves to weaken the light squares around the white kingside, and to block possible Bc1-d2-e1-g3/h4 manoeuvres

;
(2) Psychologically, White players seem to think of it as something of a desperation tactic, which Black only plays because he didn't want to play against the Flick-Knife and is casting around wildly. It encourages them in the view that they are basically already winning, which is very far from the truth! 8)
(3) It is fun to mess up White's development with a check which is only possible because White messed up Black's development with a check - there is an element of tit-for-tat about it ...