Discharge of power rails during Jetson TX2i power-down sequence

Hello. While looking through the NVIDIA Jetson TX2 Series OEM Product Design Guide, I notice that it is important to power down the local carrier card rails before the Jetson module system power is down. As described, this will require discharge circuitry. My question is as follows: is this only for the 1.8V rail or does this also pertain to all local power rails on the carrier card? The documentation states that it is mainly for the 1.8V rail but your reference carrier card circuitry shows discharge circuitry for 5V, 3V3, 1V8, etc.

Likewise, during power-up, when CARRIER_PWR_ON is driven by the TX2i, we are free to enable VDD_1V8. Should this also be a signal to enable all local power supplies on the carrier card (3V3 for example)? If it only applies to 1V8, is it OK to have the other local power supplies (which power Jetson interfaces) to be already up before CARRIER_PWR_ON goes active high?

Thanks very much,

The main purpose of powering up module earlier and powering off module later than carrier board is to prove the IOs shared by module and carrier board in stable status when module start to power off. So our suggestion would be to follow discharge design in DG, and you can self-estimate if it is workable not to do that on some rails which are not related to the shared IOs.

Thanks very much.