When do the checkpoint and header pre-sync rules start to apply?

Related to this.

It describes mostly when we already have an active chaintip; I’m interested in when we don’t and when we are doing IBD.

As far as I understand, the active chaintip is set only after we completely synchronize and find the best chain (with the most cumulative work).

1. Are checkpoint and header pre-synchronization applied only after we have set at least one active chaintip? I made my own conclusion, I don’t know if it is correct.
   • They are not applied in the first IBD and IBDs that are not first (not starting from genesis) but node have never set an active chaintip (never reached fully synchronization). For nodes that were once synced (have some out of date active chaintip, but still have), and start again after, for example, 2 months (so start IBD), checkpoints and header pre-sync apply because it has something to compare with.
2. Can this result attack for a nodes without active chaintip where attacker make low difficulty chain from genesis and send it to node? Node must store all this headers since it can not know whether will it produce chain with most work and it also can’t apply checkpoint and header pre-sync since it has not active chaintip (don’t have anything to compare with).