Currently once 80% of the validators switch to a new protocol version, the upgrade happens in 2 epochs, which may not be sufficient for some validators to upgrade their nodes. At the same time, if there is an emergency upgrade to fix some critical vulnerability, we would like to upgrade to finish as fast as possible. Therefore, it may make sense to distinguish, for every protocol upgrade, whether it is urgent or standard and have different number of epochs between when the majority of validator switch to the new version and when the new protocol kicks in.
Currently once 80% of the validators switch to a new protocol version, the upgrade happens in 2 epochs, which may not be sufficient for some validators to upgrade their nodes. At the same time, if there is an emergency upgrade to fix some critical vulnerability, we would like to upgrade to finish as fast as possible. Therefore, it may make sense to distinguish, for every protocol upgrade, whether it is urgent or standard and have different number of epochs between when the majority of validator switch to the new version and when the new protocol kicks in.