prysm-pulse/beacon-chain/core/altair/reward.go
Raul Jordan f3d6dbcc1e
Move Shared/Params Into Config/Params (#9642)
* config params into pkg

* gaz

Co-authored-by: prylabs-bulldozer[bot] <58059840+prylabs-bulldozer[bot]@users.noreply.github.com>
2021-09-21 19:59:25 +00:00

61 lines
2.3 KiB
Go

package altair
import (
"github.com/pkg/errors"
types "github.com/prysmaticlabs/eth2-types"
"github.com/prysmaticlabs/prysm/beacon-chain/core/helpers"
"github.com/prysmaticlabs/prysm/beacon-chain/state"
"github.com/prysmaticlabs/prysm/config/params"
"github.com/prysmaticlabs/prysm/math"
)
// BaseReward takes state and validator index and calculate
// individual validator's base reward.
//
// Spec code:
// def get_base_reward(state: BeaconState, index: ValidatorIndex) -> Gwei:
// """
// Return the base reward for the validator defined by ``index`` with respect to the current ``state``.
//
// Note: An optimally performing validator can earn one base reward per epoch over a long time horizon.
// This takes into account both per-epoch (e.g. attestation) and intermittent duties (e.g. block proposal
// and sync committees).
// """
// increments = state.validators[index].effective_balance // EFFECTIVE_BALANCE_INCREMENT
// return Gwei(increments * get_base_reward_per_increment(state))
func BaseReward(s state.ReadOnlyBeaconState, index types.ValidatorIndex) (uint64, error) {
totalBalance, err := helpers.TotalActiveBalance(s)
if err != nil {
return 0, errors.Wrap(err, "could not calculate active balance")
}
return BaseRewardWithTotalBalance(s, index, totalBalance)
}
// BaseRewardWithTotalBalance calculates the base reward with the provided total balance.
func BaseRewardWithTotalBalance(s state.ReadOnlyBeaconState, index types.ValidatorIndex, totalBalance uint64) (uint64, error) {
val, err := s.ValidatorAtIndexReadOnly(index)
if err != nil {
return 0, err
}
cfg := params.BeaconConfig()
increments := val.EffectiveBalance() / cfg.EffectiveBalanceIncrement
baseRewardPerInc, err := BaseRewardPerIncrement(totalBalance)
if err != nil {
return 0, err
}
return increments * baseRewardPerInc, nil
}
// BaseRewardPerIncrement of the beacon state
//
// Spec code:
// def get_base_reward_per_increment(state: BeaconState) -> Gwei:
// return Gwei(EFFECTIVE_BALANCE_INCREMENT * BASE_REWARD_FACTOR // integer_squareroot(get_total_active_balance(state)))
func BaseRewardPerIncrement(activeBalance uint64) (uint64, error) {
if activeBalance == 0 {
return 0, errors.New("active balance can't be 0")
}
cfg := params.BeaconConfig()
return cfg.EffectiveBalanceIncrement * cfg.BaseRewardFactor / math.IntegerSquareRoot(activeBalance), nil
}