@User: Also I will try to make sure there is area available in the chip center for inductors or anything else on RDL. There is more width to the chip than we had originally, so this is feasible, although I don't have the time to completely redesign the bump map now for this tapeout.

But still the routing to the inductor would be done by OpenLane??

At least for now, the RDL layer has special layers (under-bump material, for example) that are not represented in Magic and so all of the RDL part needs to be done separately. What you would probably want to do in this case is to have a glass cut probe opening where the RDL inductor would connect to metal5. Openlane would be able to route to that metal5 pad. Essentially, you'd have a placeholder macro for the inductor that includes only the part on metal5 and passivation layers.

@Tim Edwards I wish we had more time to be able to tapeout our inductor.

@Amro Tork: There will be additional shuttle runs. Google has already agreed to pay for four more in 2021. By waiting for the next one, you will have the advantage that I will have some experience with the redistribution layer back-end, which I am currently leaving as a last-minute item because it is a post-processing step and so does not need to be dealt with immediately.

@Tim Edwards no problem. We could tape out analog blocks only this time.

IIRC you can leave the harness alone, and manually route to the pads, right?

Yes

@Amro Tork you can manually route analogue without using openlane right now. Just make your layout in the user wrapper gds

Thanks @yrrapt