Skip to content

Reward Update Transition

The Reward Update Transition calculates a new RewardUpdate to apply in an epoch transition.

The environment consists of the produced blocks and the epoch state. The state is an optional reward update.

There are three transition cases, one that computes a new reward update, one that leaves the rewards update unchanged as it has not yet been applied and finally one that leaves the reward update unchanged as the transition was started too early.

The signal of the transition rule RUPD is the current slot s.

The execution of the transition role is as follows:

  • If the current reward update is empty and s is greater than the sum of the first slot of its epoch and the duration RandomnessStabilisationWindowᶜ, then a new rewards update is calculated and the state is updated.
  • If the current reward update is not nothing, i.e. a reward update has already been calculated but not yet applied, then the state is not updated.
  • If the current reward update is empty and s is less than or equal to the sum of the first slot of its epoch and the duration RandomnessStabilisationWindowᶜ, then the state is not updated.
RUpdEnv : Type
RUpdEnv = BlocksMade × EpochState


data _⊢_⇀⦇_,RUPD⦈_
  : RUpdEnv → Maybe RewardUpdate → Slot → Maybe RewardUpdate → Type where

  RUPD-Create-Reward-Update : ∀ {b es s} {ru' : RewardUpdate} →
    let
      ru' = createRUpd SlotsPerEpochᶜ b es MaxLovelaceSupplyᶜ
    in
    ∙ s > firstSlot (epoch s) + RandomnessStabilisationWindow
    ────────────────────────────────
    (b , es) ⊢ nothing ⇀⦇ s ,RUPD⦈ just ru'

  RUPD-Reward-Update-Exists : ∀ {b es s ru} →
    ────────────────────────────────
    (b , es) ⊢ just ru ⇀⦇ s ,RUPD⦈ just ru

  RUPD-Reward-Too-Early : ∀ {b es s} →
    ∙ ¬ s > firstSlot (epoch s) + RandomnessStabilisationWindow
    ────────────────────────────────
    (b , es) ⊢ nothing ⇀⦇ s ,RUPD⦈ nothing

Chain Tick Transition

The Chain Tick Transition performs some chain level upkeep. The environment consists of a set of genesis keys, and the state is the epoch specific state necessary for the NEWEPOCH transition.

Two transitions are done:

  • The NEWEPOCH transition performs any state change needed if it is the first block of a new epoch.
  • The RUPD creates the reward update if it is late enough in the epoch. Note that for every block header, either NEWEPOCH or RUPD will be the identity transition.
data _⊢_⇀⦇_,TICK⦈_
  : ⊤ → NewEpochState → Slot → NewEpochState → Type where

  TICK : ∀ {slot nes nes' ru''} →
    let open NewEpochState in
    -- TODO: Is this really how it should be?
    -- We are skipping adoptGenesisDelegs here.
    ∙ tt ⊢ nes ⇀⦇ epoch slot ,NEWEPOCH⦈ nes'
    ∙ (nes .bprev , nes .epochState) ⊢ (nes' .ru) ⇀⦇ slot ,RUPD⦈ ru''
      ────────────────────────────────
      tt ⊢ nes ⇀⦇ slot ,TICK⦈ record nes' { ru = ru'' }