status-im
/
eth2.0-specs
mirror of https://github.com/status-im/eth2.0-specs.git
2
0
Fork 0
Code Issues Projects Releases Wiki Activity

update penalty config values for hf1

This commit is contained in:
Danny Ryan 2021-02-11 14:53:29 -07:00
parent 7d715220bb
commit 2b8b0d9e2b
No known key found for this signature in database
GPG Key ID: 2765A792E42CE07A
4 changed files with 134 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
configs/mainnet/lightclient_patch.yaml
View File

@ -2,6 +2,16 @@
CONFIG_NAME: "mainnet"
# Updated penalty values
# ---------------------------------------------------------------
# 3 * 2**24) (= 50,331,648)
HF1_INACTIVITY_PENALTY_QUOTIENT: 50331648
# 2**6 (= 64)
HF1_MIN_SLASHING_PENALTY_QUOTIENT: 64
# 2
HF1_PROPORTIONAL_SLASHING_MULTIPLIER: 2
# Misc
# ---------------------------------------------------------------
# 2**10 (=1,024)

10
configs/minimal/lightclient_patch.yaml
View File

@ -2,6 +2,16 @@
CONFIG_NAME: "minimal"
# Updated penalty values
# ---------------------------------------------------------------
# 3 * 2**24) (= 50,331,648)
HF1_INACTIVITY_PENALTY_QUOTIENT: 50331648
# 2**6 (= 64)
HF1_MIN_SLASHING_PENALTY_QUOTIENT: 64
# 2
HF1_PROPORTIONAL_SLASHING_MULTIPLIER: 2
# Misc
# ---------------------------------------------------------------
# [customized]

77
specs/lightclient/beacon-chain.md
View File

@ -13,6 +13,7 @@
- [Participation rewards](#participation-rewards)
- [Misc](#misc)
- [Configuration](#configuration)
- [Updated penalty values](#updated-penalty-values)
- [Misc](#misc-1)
- [Time parameters](#time-parameters)
- [Domain types](#domain-types)
@ -34,6 +35,8 @@
- [`get_unslashed_participating_indices`](#get_unslashed_participating_indices)
- [`get_flag_deltas`](#get_flag_deltas)
- [New `get_inactivity_penalty_deltas`](#new-get_inactivity_penalty_deltas)
- [Beacon state mutators](#beacon-state-mutators)
- [New `slash_validator`](#new-slash_validator)
- [Block processing](#block-processing)
- [New `process_attestation`](#new-process_attestation)
- [New `process_deposit`](#new-process_deposit)
@ -41,6 +44,7 @@
- [Epoch processing](#epoch-processing)
- [New `process_justification_and_finalization`](#new-process_justification_and_finalization)
- [New `process_rewards_and_penalties`](#new-process_rewards_and_penalties)
- [New `process_slashings`](#new-process_slashings)
- [Sync committee updates](#sync-committee-updates)
- [Participation flags updates](#participation-flags-updates)
@ -49,7 +53,8 @@
## Introduction
This is a patch implementing the first hard fork to the beacon chain, tentatively named HF1 pending a permanent name. It has three main features:
This is a patch implementing the first hard fork to the beacon chain, tentatively named HF1 pending a permanent name.
It has three main features:
* Light client support via sync committees
* Incentive accounting reforms, reducing spec complexity
@ -97,6 +102,18 @@ The reward fractions add up to 7/8, leaving the remaining 1/8 for proposer rewar
## Configuration
### Updated penalty values
This patch updates a few configuration values to move penalty constants toward their final, maxmium security values.
*Note*: The spec does *not* override previous configuration values but instead creates new values and replaces usage throughout.
| Name | Value |
| - | - |
| `HF1_INACTIVITY_PENALTY_QUOTIENT` | `uint64(3 * 2**24)` (= 50,331,648) |
| `HF1_MIN_SLASHING_PENALTY_QUOTIENT` | `uint64(2**6)` (=64) |
| `HF1_PROPORTIONAL_SLASHING_MULTIPLIER` | `uint64(2)` |
### Misc
| Name | Value |
@ -328,7 +345,8 @@ def get_flag_deltas(state: BeaconState,
#### New `get_inactivity_penalty_deltas`
*Note*: The function `get_inactivity_penalty_deltas` is modified in the selection of matching target indices and the removal of `BASE_REWARDS_PER_EPOCH`.
*Note*: The function `get_inactivity_penalty_deltas` is modified in the selection of matching target indices
and the removal of `BASE_REWARDS_PER_EPOCH`.
```python
def get_inactivity_penalty_deltas(state: BeaconState) -> Tuple[Sequence[Gwei], Sequence[Gwei]]:
@ -348,12 +366,47 @@ def get_inactivity_penalty_deltas(state: BeaconState) -> Tuple[Sequence[Gwei], S
penalties[index] += Gwei(get_base_reward(state, index) * reward_numerator_sum // REWARD_DENOMINATOR)
if index not in matching_target_attesting_indices:
effective_balance = state.validators[index].effective_balance
penalties[index] += Gwei(effective_balance * get_finality_delay(state) // INACTIVITY_PENALTY_QUOTIENT)
penalties[index] += Gwei(
effective_balance * get_finality_delay(state)
// HF1_INACTIVITY_PENALTY_QUOTIENT
)
rewards = [Gwei(0) for _ in range(len(state.validators))]
return rewards, penalties
```
### Beacon state mutators
#### New `slash_validator`
*Note*: The function `slash_validator` is modified
with the substitution of `MIN_SLASHING_PENALTY_QUOTIENT` with `HF1_MIN_SLASHING_PENALTY_QUOTIENT`.
```python
def slash_validator(state: BeaconState,
slashed_index: ValidatorIndex,
whistleblower_index: ValidatorIndex=None) -> None:
"""
Slash the validator with index ``slashed_index``.
"""
epoch = get_current_epoch(state)
initiate_validator_exit(state, slashed_index)
validator = state.validators[slashed_index]
validator.slashed = True
validator.withdrawable_epoch = max(validator.withdrawable_epoch, Epoch(epoch + EPOCHS_PER_SLASHINGS_VECTOR))
state.slashings[epoch % EPOCHS_PER_SLASHINGS_VECTOR] += validator.effective_balance
decrease_balance(state, slashed_index, validator.effective_balance // HF1_MIN_SLASHING_PENALTY_QUOTIENT)
# Apply proposer and whistleblower rewards
proposer_index = get_beacon_proposer_index(state)
if whistleblower_index is None:
whistleblower_index = proposer_index
whistleblower_reward = Gwei(validator.effective_balance // WHISTLEBLOWER_REWARD_QUOTIENT)
proposer_reward = Gwei(whistleblower_reward // PROPOSER_REWARD_QUOTIENT)
increase_balance(state, proposer_index, proposer_reward)
increase_balance(state, whistleblower_index, Gwei(whistleblower_reward - proposer_reward))
```
### Block processing
```python
@ -576,6 +629,24 @@ def process_rewards_and_penalties(state: BeaconState) -> None:
decrease_balance(state, ValidatorIndex(index), penalties[index])
```
#### New `process_slashings`
*Note*: The function `process_slashings` is modified
with the substitution of `PROPORTIONAL_SLASHING_MULTIPLIER` with `HF1_PROPORTIONAL_SLASHING_MULTIPLIER`.
```python
def process_slashings(state: BeaconState) -> None:
epoch = get_current_epoch(state)
total_balance = get_total_active_balance(state)
adjusted_total_slashing_balance = min(sum(state.slashings) * HF1_PROPORTIONAL_SLASHING_MULTIPLIER, total_balance)
for index, validator in enumerate(state.validators):
if validator.slashed and epoch + EPOCHS_PER_SLASHINGS_VECTOR // 2 == validator.withdrawable_epoch:
increment = EFFECTIVE_BALANCE_INCREMENT # Factored out from penalty numerator to avoid uint64 overflow
penalty_numerator = validator.effective_balance // increment * adjusted_total_slashing_balance
penalty = penalty_numerator // total_balance * increment
decrease_balance(state, ValidatorIndex(index), penalty)
```
#### Sync committee updates
```python

48
tests/core/pyspec/eth2spec/test/phase0/epoch_processing/test_process_slashings.py
View File

@ -1,4 +1,4 @@
from eth2spec.test.context import spec_state_test, with_all_phases
from eth2spec.test.context import spec_state_test, with_all_phases, is_post_lightclient_patch
from eth2spec.test.helpers.epoch_processing import (
run_epoch_processing_with, run_epoch_processing_to
)
@ -23,12 +23,19 @@ def slash_validators(spec, state, indices, out_epochs):
] = total_slashed_balance
def get_slashing_multipler(spec):
if is_post_lightclient_patch(spec):
return spec.HF1_PROPORTIONAL_SLASHING_MULTIPLIER
else:
return spec.PROPORTIONAL_SLASHING_MULTIPLIER
@with_all_phases
@spec_state_test
def test_max_penalties(spec, state):
# Slashed count to ensure that enough validators are slashed to induce maximum penalties
slashed_count = min(
(len(state.validators) // spec.PROPORTIONAL_SLASHING_MULTIPLIER) + 1,
(len(state.validators) // get_slashing_multipler(spec)) + 1,
# Can't slash more than validator count!
len(state.validators)
)
@ -40,7 +47,7 @@ def test_max_penalties(spec, state):
total_balance = spec.get_total_active_balance(state)
total_penalties = sum(state.slashings)
assert total_balance // spec.PROPORTIONAL_SLASHING_MULTIPLIER <= total_penalties
assert total_balance // get_slashing_multipler(spec) <= total_penalties
yield from run_process_slashings(spec, state)
@ -50,7 +57,30 @@ def test_max_penalties(spec, state):
@with_all_phases
@spec_state_test
def test_small_penalty(spec, state):
def test_low_penalty(spec, state):
# Slashed count is one tenth of validator set
slashed_count = (len(state.validators) // 10) + 1
out_epoch = spec.get_current_epoch(state) + (spec.EPOCHS_PER_SLASHINGS_VECTOR // 2)
slashed_indices = list(range(slashed_count))
slash_validators(spec, state, slashed_indices, [out_epoch] * slashed_count)
pre_state = state.copy()
yield from run_process_slashings(spec, state)
for i in slashed_indices:
assert 0 < state.balances[i] < pre_state.balances[i]
@with_all_phases
@spec_state_test
def test_minimal_penalty(spec, state):
#
# When very few slashings, the resulting slashing penalty gets rounded down
# to zero so the result of `process_slashings` is null
#
# Just the bare minimum for this one validator
state.balances[0] = state.validators[0].effective_balance = spec.EJECTION_BALANCE
# All the other validators get the maximum.
@ -74,11 +104,13 @@ def test_small_penalty(spec, state):
expected_penalty = (
state.validators[0].effective_balance // spec.EFFECTIVE_BALANCE_INCREMENT
* (3 * total_penalties)
* (get_slashing_multipler(spec) * total_penalties)
// total_balance
* spec.EFFECTIVE_BALANCE_INCREMENT
)
assert state.balances[0] == pre_slash_balances[0] - expected_penalty
assert expected_penalty == 0
assert state.balances[0] == pre_slash_balances[0]
@with_all_phases
@ -96,7 +128,7 @@ def test_scaled_penalties(spec, state):
state.slashings[5] = base + (incr * 6)
state.slashings[spec.EPOCHS_PER_SLASHINGS_VECTOR - 1] = base + (incr * 7)
slashed_count = len(state.validators) // (spec.PROPORTIONAL_SLASHING_MULTIPLIER + 1)
slashed_count = len(state.validators) // (get_slashing_multipler(spec) + 1)
assert slashed_count > 10
@ -134,7 +166,7 @@ def test_scaled_penalties(spec, state):
v = state.validators[i]
expected_penalty = (
v.effective_balance // spec.EFFECTIVE_BALANCE_INCREMENT
* (spec.PROPORTIONAL_SLASHING_MULTIPLIER * total_penalties)
* (get_slashing_multipler(spec) * total_penalties)
// (total_balance)
* spec.EFFECTIVE_BALANCE_INCREMENT
)