Apologies for not looking into this thread to often, I've been on holiday.
I am inclined to go with phth's proposal as the winning entry. I like the creativity of the approach. I would like to see a synthesis of the proposed starting chiral acid chloride though...
I will share with you my own suggestion. I spent some time trying to come up with a chiral pool synthesis from pyrogluctamic acid, but failed to find something good. Eventually I settled on this as the shortest route. It is an enantioselective synthesis using Noyori asymmetric hydrogenation for stereocontrol. Not very inventive, but it is short.
1. Double addition of of lithiated hexyne to succinic anhydride at low temp (diketone obtained on aqueous workup)
2. Lindlar reduction [Edit: Alkenes formed in this step should be Z, not E]
3. Double Noyori reduction (the chirality sense at both asymmetric centres is the same)
4. Double tosylation
5. Double SN2 displacement by the chiral allylic amine (itself prepared in 3 steps from 3-octen-2-one via Noyori chemistry).
6. Grubbs metathesis
7. Alkene hydrogenation
Longest linear sequence is 7 steps.