2021-08-06 00:47:31 +00:00
|
|
|
//! Tests for API behaviour across fork boundaries.
|
|
|
|
use crate::common::*;
|
|
|
|
use beacon_chain::{test_utils::RelativeSyncCommittee, StateSkipConfig};
|
2023-01-20 23:39:59 +00:00
|
|
|
use eth2::types::{IndexedErrorMessage, StateId, SyncSubcommittee};
|
|
|
|
use types::{Address, ChainSpec, Epoch, EthSpec, MinimalEthSpec, Slot};
|
2021-08-06 00:47:31 +00:00
|
|
|
|
|
|
|
type E = MinimalEthSpec;
|
|
|
|
|
|
|
|
fn altair_spec(altair_fork_epoch: Epoch) -> ChainSpec {
|
|
|
|
let mut spec = E::default_spec();
|
|
|
|
spec.altair_fork_epoch = Some(altair_fork_epoch);
|
|
|
|
spec
|
|
|
|
}
|
|
|
|
|
2023-01-20 23:39:59 +00:00
|
|
|
fn capella_spec(capella_fork_epoch: Epoch) -> ChainSpec {
|
|
|
|
let mut spec = E::default_spec();
|
|
|
|
spec.altair_fork_epoch = Some(Epoch::new(0));
|
|
|
|
spec.bellatrix_fork_epoch = Some(Epoch::new(0));
|
|
|
|
spec.capella_fork_epoch = Some(capella_fork_epoch);
|
|
|
|
spec
|
|
|
|
}
|
|
|
|
|
2021-08-06 00:47:31 +00:00
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
|
|
async fn sync_committee_duties_across_fork() {
|
|
|
|
let validator_count = E::sync_committee_size();
|
|
|
|
let fork_epoch = Epoch::new(8);
|
|
|
|
let spec = altair_spec(fork_epoch);
|
2021-10-11 02:45:06 +00:00
|
|
|
let tester = InteractiveTester::<E>::new(Some(spec.clone()), validator_count).await;
|
2021-08-06 00:47:31 +00:00
|
|
|
let harness = &tester.harness;
|
|
|
|
let client = &tester.client;
|
|
|
|
|
|
|
|
let all_validators = harness.get_all_validators();
|
|
|
|
let all_validators_u64 = all_validators.iter().map(|x| *x as u64).collect::<Vec<_>>();
|
|
|
|
|
|
|
|
assert_eq!(harness.get_current_slot(), 0);
|
|
|
|
|
|
|
|
// Prior to the fork the endpoint should return an empty vec.
|
|
|
|
let early_duties = client
|
|
|
|
.post_validator_duties_sync(fork_epoch - 1, &all_validators_u64)
|
|
|
|
.await
|
|
|
|
.unwrap()
|
|
|
|
.data;
|
|
|
|
assert!(early_duties.is_empty());
|
|
|
|
|
|
|
|
// If there's a skip slot at the fork slot, the endpoint should return duties, even
|
|
|
|
// though the head state hasn't transitioned yet.
|
|
|
|
let fork_slot = fork_epoch.start_slot(E::slots_per_epoch());
|
|
|
|
let (genesis_state, genesis_state_root) = harness.get_current_state_and_root();
|
|
|
|
let (_, state) = harness
|
|
|
|
.add_attested_block_at_slot(
|
|
|
|
fork_slot - 1,
|
|
|
|
genesis_state,
|
|
|
|
genesis_state_root,
|
|
|
|
&all_validators,
|
|
|
|
)
|
Use async code when interacting with EL (#3244)
## Overview
This rather extensive PR achieves two primary goals:
1. Uses the finalized/justified checkpoints of fork choice (FC), rather than that of the head state.
2. Refactors fork choice, block production and block processing to `async` functions.
Additionally, it achieves:
- Concurrent forkchoice updates to the EL and cache pruning after a new head is selected.
- Concurrent "block packing" (attestations, etc) and execution payload retrieval during block production.
- Concurrent per-block-processing and execution payload verification during block processing.
- The `Arc`-ification of `SignedBeaconBlock` during block processing (it's never mutated, so why not?):
- I had to do this to deal with sending blocks into spawned tasks.
- Previously we were cloning the beacon block at least 2 times during each block processing, these clones are either removed or turned into cheaper `Arc` clones.
- We were also `Box`-ing and un-`Box`-ing beacon blocks as they moved throughout the networking crate. This is not a big deal, but it's nice to avoid shifting things between the stack and heap.
- Avoids cloning *all the blocks* in *every chain segment* during sync.
- It also has the potential to clean up our code where we need to pass an *owned* block around so we can send it back in the case of an error (I didn't do much of this, my PR is already big enough :sweat_smile:)
- The `BeaconChain::HeadSafetyStatus` struct was removed. It was an old relic from prior merge specs.
For motivation for this change, see https://github.com/sigp/lighthouse/pull/3244#issuecomment-1160963273
## Changes to `canonical_head` and `fork_choice`
Previously, the `BeaconChain` had two separate fields:
```
canonical_head: RwLock<Snapshot>,
fork_choice: RwLock<BeaconForkChoice>
```
Now, we have grouped these values under a single struct:
```
canonical_head: CanonicalHead {
cached_head: RwLock<Arc<Snapshot>>,
fork_choice: RwLock<BeaconForkChoice>
}
```
Apart from ergonomics, the only *actual* change here is wrapping the canonical head snapshot in an `Arc`. This means that we no longer need to hold the `cached_head` (`canonical_head`, in old terms) lock when we want to pull some values from it. This was done to avoid deadlock risks by preventing functions from acquiring (and holding) the `cached_head` and `fork_choice` locks simultaneously.
## Breaking Changes
### The `state` (root) field in the `finalized_checkpoint` SSE event
Consider the scenario where epoch `n` is just finalized, but `start_slot(n)` is skipped. There are two state roots we might in the `finalized_checkpoint` SSE event:
1. The state root of the finalized block, which is `get_block(finalized_checkpoint.root).state_root`.
4. The state root at slot of `start_slot(n)`, which would be the state from (1), but "skipped forward" through any skip slots.
Previously, Lighthouse would choose (2). However, we can see that when [Teku generates that event](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/beaconrestapi/src/main/java/tech/pegasys/teku/beaconrestapi/handlers/v1/events/EventSubscriptionManager.java#L171-L182) it uses [`getStateRootFromBlockRoot`](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/provider/src/main/java/tech/pegasys/teku/api/ChainDataProvider.java#L336-L341) which uses (1).
I have switched Lighthouse from (2) to (1). I think it's a somewhat arbitrary choice between the two, where (1) is easier to compute and is consistent with Teku.
## Notes for Reviewers
I've renamed `BeaconChain::fork_choice` to `BeaconChain::recompute_head`. Doing this helped ensure I broke all previous uses of fork choice and I also find it more descriptive. It describes an action and can't be confused with trying to get a reference to the `ForkChoice` struct.
I've changed the ordering of SSE events when a block is received. It used to be `[block, finalized, head]` and now it's `[block, head, finalized]`. It was easier this way and I don't think we were making any promises about SSE event ordering so it's not "breaking".
I've made it so fork choice will run when it's first constructed. I did this because I wanted to have a cached version of the last call to `get_head`. Ensuring `get_head` has been run *at least once* means that the cached values doesn't need to wrapped in an `Option`. This was fairly simple, it just involved passing a `slot` to the constructor so it knows *when* it's being run. When loading a fork choice from the store and a slot clock isn't handy I've just used the `slot` that was saved in the `fork_choice_store`. That seems like it would be a faithful representation of the slot when we saved it.
I added the `genesis_time: u64` to the `BeaconChain`. It's small, constant and nice to have around.
Since we're using FC for the fin/just checkpoints, we no longer get the `0x00..00` roots at genesis. You can see I had to remove a work-around in `ef-tests` here: b56be3bc2. I can't find any reason why this would be an issue, if anything I think it'll be better since the genesis-alias has caught us out a few times (0x00..00 isn't actually a real root). Edit: I did find a case where the `network` expected the 0x00..00 alias and patched it here: 3f26ac3e2.
You'll notice a lot of changes in tests. Generally, tests should be functionally equivalent. Here are the things creating the most diff-noise in tests:
- Changing tests to be `tokio::async` tests.
- Adding `.await` to fork choice, block processing and block production functions.
- Refactor of the `canonical_head` "API" provided by the `BeaconChain`. E.g., `chain.canonical_head.cached_head()` instead of `chain.canonical_head.read()`.
- Wrapping `SignedBeaconBlock` in an `Arc`.
- In the `beacon_chain/tests/block_verification`, we can't use the `lazy_static` `CHAIN_SEGMENT` variable anymore since it's generated with an async function. We just generate it in each test, not so efficient but hopefully insignificant.
I had to disable `rayon` concurrent tests in the `fork_choice` tests. This is because the use of `rayon` and `block_on` was causing a panic.
Co-authored-by: Mac L <mjladson@pm.me>
2022-07-03 05:36:50 +00:00
|
|
|
.await
|
2021-08-06 00:47:31 +00:00
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
harness.advance_slot();
|
|
|
|
assert_eq!(harness.get_current_slot(), fork_slot);
|
|
|
|
|
|
|
|
let sync_duties = client
|
|
|
|
.post_validator_duties_sync(fork_epoch, &all_validators_u64)
|
|
|
|
.await
|
|
|
|
.unwrap()
|
|
|
|
.data;
|
|
|
|
assert_eq!(sync_duties.len(), E::sync_committee_size());
|
|
|
|
|
|
|
|
// After applying a block at the fork slot the duties should remain unchanged.
|
|
|
|
let state_root = state.canonical_root();
|
|
|
|
harness
|
|
|
|
.add_attested_block_at_slot(fork_slot, state, state_root, &all_validators)
|
Use async code when interacting with EL (#3244)
## Overview
This rather extensive PR achieves two primary goals:
1. Uses the finalized/justified checkpoints of fork choice (FC), rather than that of the head state.
2. Refactors fork choice, block production and block processing to `async` functions.
Additionally, it achieves:
- Concurrent forkchoice updates to the EL and cache pruning after a new head is selected.
- Concurrent "block packing" (attestations, etc) and execution payload retrieval during block production.
- Concurrent per-block-processing and execution payload verification during block processing.
- The `Arc`-ification of `SignedBeaconBlock` during block processing (it's never mutated, so why not?):
- I had to do this to deal with sending blocks into spawned tasks.
- Previously we were cloning the beacon block at least 2 times during each block processing, these clones are either removed or turned into cheaper `Arc` clones.
- We were also `Box`-ing and un-`Box`-ing beacon blocks as they moved throughout the networking crate. This is not a big deal, but it's nice to avoid shifting things between the stack and heap.
- Avoids cloning *all the blocks* in *every chain segment* during sync.
- It also has the potential to clean up our code where we need to pass an *owned* block around so we can send it back in the case of an error (I didn't do much of this, my PR is already big enough :sweat_smile:)
- The `BeaconChain::HeadSafetyStatus` struct was removed. It was an old relic from prior merge specs.
For motivation for this change, see https://github.com/sigp/lighthouse/pull/3244#issuecomment-1160963273
## Changes to `canonical_head` and `fork_choice`
Previously, the `BeaconChain` had two separate fields:
```
canonical_head: RwLock<Snapshot>,
fork_choice: RwLock<BeaconForkChoice>
```
Now, we have grouped these values under a single struct:
```
canonical_head: CanonicalHead {
cached_head: RwLock<Arc<Snapshot>>,
fork_choice: RwLock<BeaconForkChoice>
}
```
Apart from ergonomics, the only *actual* change here is wrapping the canonical head snapshot in an `Arc`. This means that we no longer need to hold the `cached_head` (`canonical_head`, in old terms) lock when we want to pull some values from it. This was done to avoid deadlock risks by preventing functions from acquiring (and holding) the `cached_head` and `fork_choice` locks simultaneously.
## Breaking Changes
### The `state` (root) field in the `finalized_checkpoint` SSE event
Consider the scenario where epoch `n` is just finalized, but `start_slot(n)` is skipped. There are two state roots we might in the `finalized_checkpoint` SSE event:
1. The state root of the finalized block, which is `get_block(finalized_checkpoint.root).state_root`.
4. The state root at slot of `start_slot(n)`, which would be the state from (1), but "skipped forward" through any skip slots.
Previously, Lighthouse would choose (2). However, we can see that when [Teku generates that event](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/beaconrestapi/src/main/java/tech/pegasys/teku/beaconrestapi/handlers/v1/events/EventSubscriptionManager.java#L171-L182) it uses [`getStateRootFromBlockRoot`](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/provider/src/main/java/tech/pegasys/teku/api/ChainDataProvider.java#L336-L341) which uses (1).
I have switched Lighthouse from (2) to (1). I think it's a somewhat arbitrary choice between the two, where (1) is easier to compute and is consistent with Teku.
## Notes for Reviewers
I've renamed `BeaconChain::fork_choice` to `BeaconChain::recompute_head`. Doing this helped ensure I broke all previous uses of fork choice and I also find it more descriptive. It describes an action and can't be confused with trying to get a reference to the `ForkChoice` struct.
I've changed the ordering of SSE events when a block is received. It used to be `[block, finalized, head]` and now it's `[block, head, finalized]`. It was easier this way and I don't think we were making any promises about SSE event ordering so it's not "breaking".
I've made it so fork choice will run when it's first constructed. I did this because I wanted to have a cached version of the last call to `get_head`. Ensuring `get_head` has been run *at least once* means that the cached values doesn't need to wrapped in an `Option`. This was fairly simple, it just involved passing a `slot` to the constructor so it knows *when* it's being run. When loading a fork choice from the store and a slot clock isn't handy I've just used the `slot` that was saved in the `fork_choice_store`. That seems like it would be a faithful representation of the slot when we saved it.
I added the `genesis_time: u64` to the `BeaconChain`. It's small, constant and nice to have around.
Since we're using FC for the fin/just checkpoints, we no longer get the `0x00..00` roots at genesis. You can see I had to remove a work-around in `ef-tests` here: b56be3bc2. I can't find any reason why this would be an issue, if anything I think it'll be better since the genesis-alias has caught us out a few times (0x00..00 isn't actually a real root). Edit: I did find a case where the `network` expected the 0x00..00 alias and patched it here: 3f26ac3e2.
You'll notice a lot of changes in tests. Generally, tests should be functionally equivalent. Here are the things creating the most diff-noise in tests:
- Changing tests to be `tokio::async` tests.
- Adding `.await` to fork choice, block processing and block production functions.
- Refactor of the `canonical_head` "API" provided by the `BeaconChain`. E.g., `chain.canonical_head.cached_head()` instead of `chain.canonical_head.read()`.
- Wrapping `SignedBeaconBlock` in an `Arc`.
- In the `beacon_chain/tests/block_verification`, we can't use the `lazy_static` `CHAIN_SEGMENT` variable anymore since it's generated with an async function. We just generate it in each test, not so efficient but hopefully insignificant.
I had to disable `rayon` concurrent tests in the `fork_choice` tests. This is because the use of `rayon` and `block_on` was causing a panic.
Co-authored-by: Mac L <mjladson@pm.me>
2022-07-03 05:36:50 +00:00
|
|
|
.await
|
2021-08-06 00:47:31 +00:00
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
assert_eq!(
|
|
|
|
client
|
|
|
|
.post_validator_duties_sync(fork_epoch, &all_validators_u64)
|
|
|
|
.await
|
|
|
|
.unwrap()
|
|
|
|
.data,
|
|
|
|
sync_duties
|
|
|
|
);
|
|
|
|
|
|
|
|
// Sync duties should also be available for the next period.
|
|
|
|
let current_period = fork_epoch.sync_committee_period(&spec).unwrap();
|
|
|
|
let next_period_epoch = spec.epochs_per_sync_committee_period * (current_period + 1);
|
|
|
|
|
|
|
|
let next_period_duties = client
|
|
|
|
.post_validator_duties_sync(next_period_epoch, &all_validators_u64)
|
|
|
|
.await
|
|
|
|
.unwrap()
|
|
|
|
.data;
|
|
|
|
assert_eq!(next_period_duties.len(), E::sync_committee_size());
|
|
|
|
|
|
|
|
// Sync duties should *not* be available for the period after the next period.
|
|
|
|
// We expect a 400 (bad request) response.
|
|
|
|
let next_next_period_epoch = spec.epochs_per_sync_committee_period * (current_period + 2);
|
|
|
|
assert_eq!(
|
|
|
|
client
|
|
|
|
.post_validator_duties_sync(next_next_period_epoch, &all_validators_u64)
|
|
|
|
.await
|
|
|
|
.unwrap_err()
|
|
|
|
.status()
|
|
|
|
.unwrap(),
|
|
|
|
400
|
|
|
|
);
|
|
|
|
}
|
|
|
|
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
|
|
async fn attestations_across_fork_with_skip_slots() {
|
|
|
|
let validator_count = E::sync_committee_size();
|
|
|
|
let fork_epoch = Epoch::new(8);
|
|
|
|
let spec = altair_spec(fork_epoch);
|
2021-10-11 02:45:06 +00:00
|
|
|
let tester = InteractiveTester::<E>::new(Some(spec.clone()), validator_count).await;
|
2021-08-06 00:47:31 +00:00
|
|
|
let harness = &tester.harness;
|
|
|
|
let client = &tester.client;
|
|
|
|
|
|
|
|
let all_validators = harness.get_all_validators();
|
|
|
|
|
|
|
|
let fork_slot = fork_epoch.start_slot(E::slots_per_epoch());
|
|
|
|
let fork_state = harness
|
|
|
|
.chain
|
|
|
|
.state_at_slot(fork_slot, StateSkipConfig::WithStateRoots)
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
harness.set_current_slot(fork_slot);
|
|
|
|
|
|
|
|
let attestations = harness.make_attestations(
|
|
|
|
&all_validators,
|
|
|
|
&fork_state,
|
|
|
|
fork_state.canonical_root(),
|
|
|
|
(*fork_state.get_block_root(fork_slot - 1).unwrap()).into(),
|
|
|
|
fork_slot,
|
|
|
|
);
|
|
|
|
|
|
|
|
let unaggregated_attestations = attestations
|
|
|
|
.iter()
|
|
|
|
.flat_map(|(atts, _)| atts.iter().map(|(att, _)| att.clone()))
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
|
|
|
|
assert!(!unaggregated_attestations.is_empty());
|
|
|
|
client
|
|
|
|
.post_beacon_pool_attestations(&unaggregated_attestations)
|
|
|
|
.await
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
let signed_aggregates = attestations
|
|
|
|
.into_iter()
|
|
|
|
.filter_map(|(_, op_aggregate)| op_aggregate)
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
assert!(!signed_aggregates.is_empty());
|
|
|
|
|
|
|
|
client
|
|
|
|
.post_validator_aggregate_and_proof(&signed_aggregates)
|
|
|
|
.await
|
|
|
|
.unwrap();
|
|
|
|
}
|
|
|
|
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
|
|
async fn sync_contributions_across_fork_with_skip_slots() {
|
|
|
|
let validator_count = E::sync_committee_size();
|
|
|
|
let fork_epoch = Epoch::new(8);
|
|
|
|
let spec = altair_spec(fork_epoch);
|
2021-10-11 02:45:06 +00:00
|
|
|
let tester = InteractiveTester::<E>::new(Some(spec.clone()), validator_count).await;
|
2021-08-06 00:47:31 +00:00
|
|
|
let harness = &tester.harness;
|
|
|
|
let client = &tester.client;
|
|
|
|
|
|
|
|
let fork_slot = fork_epoch.start_slot(E::slots_per_epoch());
|
|
|
|
let fork_state = harness
|
|
|
|
.chain
|
|
|
|
.state_at_slot(fork_slot, StateSkipConfig::WithStateRoots)
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
harness.set_current_slot(fork_slot);
|
|
|
|
|
|
|
|
let sync_messages = harness.make_sync_contributions(
|
|
|
|
&fork_state,
|
|
|
|
*fork_state.get_block_root(fork_slot - 1).unwrap(),
|
|
|
|
fork_slot,
|
|
|
|
RelativeSyncCommittee::Current,
|
|
|
|
);
|
|
|
|
|
|
|
|
let sync_committee_messages = sync_messages
|
|
|
|
.iter()
|
|
|
|
.flat_map(|(messages, _)| messages.iter().map(|(message, _subnet)| message.clone()))
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
assert!(!sync_committee_messages.is_empty());
|
|
|
|
|
|
|
|
client
|
|
|
|
.post_beacon_pool_sync_committee_signatures(&sync_committee_messages)
|
|
|
|
.await
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
let signed_contributions = sync_messages
|
|
|
|
.into_iter()
|
|
|
|
.filter_map(|(_, op_aggregate)| op_aggregate)
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
assert!(!signed_contributions.is_empty());
|
|
|
|
|
|
|
|
client
|
|
|
|
.post_validator_contribution_and_proofs(&signed_contributions)
|
|
|
|
.await
|
|
|
|
.unwrap();
|
|
|
|
}
|
|
|
|
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
|
|
async fn sync_committee_indices_across_fork() {
|
|
|
|
let validator_count = E::sync_committee_size();
|
|
|
|
let fork_epoch = Epoch::new(8);
|
|
|
|
let spec = altair_spec(fork_epoch);
|
2021-10-11 02:45:06 +00:00
|
|
|
let tester = InteractiveTester::<E>::new(Some(spec.clone()), validator_count).await;
|
2021-08-06 00:47:31 +00:00
|
|
|
let harness = &tester.harness;
|
|
|
|
let client = &tester.client;
|
|
|
|
|
|
|
|
let all_validators = harness.get_all_validators();
|
|
|
|
|
|
|
|
// Flatten subcommittees into a single vec.
|
|
|
|
let flatten = |subcommittees: &[SyncSubcommittee]| -> Vec<u64> {
|
|
|
|
subcommittees
|
|
|
|
.iter()
|
|
|
|
.flat_map(|sub| sub.indices.iter().copied())
|
|
|
|
.collect()
|
|
|
|
};
|
|
|
|
|
|
|
|
// Prior to the fork the `sync_committees` endpoint should return a 400 error.
|
|
|
|
assert_eq!(
|
|
|
|
client
|
|
|
|
.get_beacon_states_sync_committees(StateId::Slot(Slot::new(0)), None)
|
|
|
|
.await
|
|
|
|
.unwrap_err()
|
|
|
|
.status()
|
|
|
|
.unwrap(),
|
|
|
|
400
|
|
|
|
);
|
|
|
|
assert_eq!(
|
|
|
|
client
|
|
|
|
.get_beacon_states_sync_committees(StateId::Head, Some(Epoch::new(0)))
|
|
|
|
.await
|
|
|
|
.unwrap_err()
|
|
|
|
.status()
|
|
|
|
.unwrap(),
|
|
|
|
400
|
|
|
|
);
|
|
|
|
|
|
|
|
// If there's a skip slot at the fork slot, the endpoint will return a 400 until a block is
|
|
|
|
// applied.
|
|
|
|
let fork_slot = fork_epoch.start_slot(E::slots_per_epoch());
|
|
|
|
let (genesis_state, genesis_state_root) = harness.get_current_state_and_root();
|
|
|
|
let (_, state) = harness
|
|
|
|
.add_attested_block_at_slot(
|
|
|
|
fork_slot - 1,
|
|
|
|
genesis_state,
|
|
|
|
genesis_state_root,
|
|
|
|
&all_validators,
|
|
|
|
)
|
Use async code when interacting with EL (#3244)
## Overview
This rather extensive PR achieves two primary goals:
1. Uses the finalized/justified checkpoints of fork choice (FC), rather than that of the head state.
2. Refactors fork choice, block production and block processing to `async` functions.
Additionally, it achieves:
- Concurrent forkchoice updates to the EL and cache pruning after a new head is selected.
- Concurrent "block packing" (attestations, etc) and execution payload retrieval during block production.
- Concurrent per-block-processing and execution payload verification during block processing.
- The `Arc`-ification of `SignedBeaconBlock` during block processing (it's never mutated, so why not?):
- I had to do this to deal with sending blocks into spawned tasks.
- Previously we were cloning the beacon block at least 2 times during each block processing, these clones are either removed or turned into cheaper `Arc` clones.
- We were also `Box`-ing and un-`Box`-ing beacon blocks as they moved throughout the networking crate. This is not a big deal, but it's nice to avoid shifting things between the stack and heap.
- Avoids cloning *all the blocks* in *every chain segment* during sync.
- It also has the potential to clean up our code where we need to pass an *owned* block around so we can send it back in the case of an error (I didn't do much of this, my PR is already big enough :sweat_smile:)
- The `BeaconChain::HeadSafetyStatus` struct was removed. It was an old relic from prior merge specs.
For motivation for this change, see https://github.com/sigp/lighthouse/pull/3244#issuecomment-1160963273
## Changes to `canonical_head` and `fork_choice`
Previously, the `BeaconChain` had two separate fields:
```
canonical_head: RwLock<Snapshot>,
fork_choice: RwLock<BeaconForkChoice>
```
Now, we have grouped these values under a single struct:
```
canonical_head: CanonicalHead {
cached_head: RwLock<Arc<Snapshot>>,
fork_choice: RwLock<BeaconForkChoice>
}
```
Apart from ergonomics, the only *actual* change here is wrapping the canonical head snapshot in an `Arc`. This means that we no longer need to hold the `cached_head` (`canonical_head`, in old terms) lock when we want to pull some values from it. This was done to avoid deadlock risks by preventing functions from acquiring (and holding) the `cached_head` and `fork_choice` locks simultaneously.
## Breaking Changes
### The `state` (root) field in the `finalized_checkpoint` SSE event
Consider the scenario where epoch `n` is just finalized, but `start_slot(n)` is skipped. There are two state roots we might in the `finalized_checkpoint` SSE event:
1. The state root of the finalized block, which is `get_block(finalized_checkpoint.root).state_root`.
4. The state root at slot of `start_slot(n)`, which would be the state from (1), but "skipped forward" through any skip slots.
Previously, Lighthouse would choose (2). However, we can see that when [Teku generates that event](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/beaconrestapi/src/main/java/tech/pegasys/teku/beaconrestapi/handlers/v1/events/EventSubscriptionManager.java#L171-L182) it uses [`getStateRootFromBlockRoot`](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/provider/src/main/java/tech/pegasys/teku/api/ChainDataProvider.java#L336-L341) which uses (1).
I have switched Lighthouse from (2) to (1). I think it's a somewhat arbitrary choice between the two, where (1) is easier to compute and is consistent with Teku.
## Notes for Reviewers
I've renamed `BeaconChain::fork_choice` to `BeaconChain::recompute_head`. Doing this helped ensure I broke all previous uses of fork choice and I also find it more descriptive. It describes an action and can't be confused with trying to get a reference to the `ForkChoice` struct.
I've changed the ordering of SSE events when a block is received. It used to be `[block, finalized, head]` and now it's `[block, head, finalized]`. It was easier this way and I don't think we were making any promises about SSE event ordering so it's not "breaking".
I've made it so fork choice will run when it's first constructed. I did this because I wanted to have a cached version of the last call to `get_head`. Ensuring `get_head` has been run *at least once* means that the cached values doesn't need to wrapped in an `Option`. This was fairly simple, it just involved passing a `slot` to the constructor so it knows *when* it's being run. When loading a fork choice from the store and a slot clock isn't handy I've just used the `slot` that was saved in the `fork_choice_store`. That seems like it would be a faithful representation of the slot when we saved it.
I added the `genesis_time: u64` to the `BeaconChain`. It's small, constant and nice to have around.
Since we're using FC for the fin/just checkpoints, we no longer get the `0x00..00` roots at genesis. You can see I had to remove a work-around in `ef-tests` here: b56be3bc2. I can't find any reason why this would be an issue, if anything I think it'll be better since the genesis-alias has caught us out a few times (0x00..00 isn't actually a real root). Edit: I did find a case where the `network` expected the 0x00..00 alias and patched it here: 3f26ac3e2.
You'll notice a lot of changes in tests. Generally, tests should be functionally equivalent. Here are the things creating the most diff-noise in tests:
- Changing tests to be `tokio::async` tests.
- Adding `.await` to fork choice, block processing and block production functions.
- Refactor of the `canonical_head` "API" provided by the `BeaconChain`. E.g., `chain.canonical_head.cached_head()` instead of `chain.canonical_head.read()`.
- Wrapping `SignedBeaconBlock` in an `Arc`.
- In the `beacon_chain/tests/block_verification`, we can't use the `lazy_static` `CHAIN_SEGMENT` variable anymore since it's generated with an async function. We just generate it in each test, not so efficient but hopefully insignificant.
I had to disable `rayon` concurrent tests in the `fork_choice` tests. This is because the use of `rayon` and `block_on` was causing a panic.
Co-authored-by: Mac L <mjladson@pm.me>
2022-07-03 05:36:50 +00:00
|
|
|
.await
|
2021-08-06 00:47:31 +00:00
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
harness.advance_slot();
|
|
|
|
assert_eq!(harness.get_current_slot(), fork_slot);
|
|
|
|
|
|
|
|
// Using the head state must fail.
|
|
|
|
assert_eq!(
|
|
|
|
client
|
|
|
|
.get_beacon_states_sync_committees(StateId::Head, Some(fork_epoch))
|
|
|
|
.await
|
|
|
|
.unwrap_err()
|
|
|
|
.status()
|
|
|
|
.unwrap(),
|
|
|
|
400
|
|
|
|
);
|
|
|
|
|
|
|
|
// In theory we could do a state advance and make this work, but to keep things simple I've
|
|
|
|
// avoided doing that for now.
|
|
|
|
assert_eq!(
|
|
|
|
client
|
|
|
|
.get_beacon_states_sync_committees(StateId::Slot(fork_slot), None)
|
|
|
|
.await
|
|
|
|
.unwrap_err()
|
|
|
|
.status()
|
|
|
|
.unwrap(),
|
|
|
|
400
|
|
|
|
);
|
|
|
|
|
|
|
|
// Once the head is updated it should be useable for requests, including in the next sync
|
|
|
|
// committee period.
|
|
|
|
let state_root = state.canonical_root();
|
|
|
|
harness
|
|
|
|
.add_attested_block_at_slot(fork_slot + 1, state, state_root, &all_validators)
|
Use async code when interacting with EL (#3244)
## Overview
This rather extensive PR achieves two primary goals:
1. Uses the finalized/justified checkpoints of fork choice (FC), rather than that of the head state.
2. Refactors fork choice, block production and block processing to `async` functions.
Additionally, it achieves:
- Concurrent forkchoice updates to the EL and cache pruning after a new head is selected.
- Concurrent "block packing" (attestations, etc) and execution payload retrieval during block production.
- Concurrent per-block-processing and execution payload verification during block processing.
- The `Arc`-ification of `SignedBeaconBlock` during block processing (it's never mutated, so why not?):
- I had to do this to deal with sending blocks into spawned tasks.
- Previously we were cloning the beacon block at least 2 times during each block processing, these clones are either removed or turned into cheaper `Arc` clones.
- We were also `Box`-ing and un-`Box`-ing beacon blocks as they moved throughout the networking crate. This is not a big deal, but it's nice to avoid shifting things between the stack and heap.
- Avoids cloning *all the blocks* in *every chain segment* during sync.
- It also has the potential to clean up our code where we need to pass an *owned* block around so we can send it back in the case of an error (I didn't do much of this, my PR is already big enough :sweat_smile:)
- The `BeaconChain::HeadSafetyStatus` struct was removed. It was an old relic from prior merge specs.
For motivation for this change, see https://github.com/sigp/lighthouse/pull/3244#issuecomment-1160963273
## Changes to `canonical_head` and `fork_choice`
Previously, the `BeaconChain` had two separate fields:
```
canonical_head: RwLock<Snapshot>,
fork_choice: RwLock<BeaconForkChoice>
```
Now, we have grouped these values under a single struct:
```
canonical_head: CanonicalHead {
cached_head: RwLock<Arc<Snapshot>>,
fork_choice: RwLock<BeaconForkChoice>
}
```
Apart from ergonomics, the only *actual* change here is wrapping the canonical head snapshot in an `Arc`. This means that we no longer need to hold the `cached_head` (`canonical_head`, in old terms) lock when we want to pull some values from it. This was done to avoid deadlock risks by preventing functions from acquiring (and holding) the `cached_head` and `fork_choice` locks simultaneously.
## Breaking Changes
### The `state` (root) field in the `finalized_checkpoint` SSE event
Consider the scenario where epoch `n` is just finalized, but `start_slot(n)` is skipped. There are two state roots we might in the `finalized_checkpoint` SSE event:
1. The state root of the finalized block, which is `get_block(finalized_checkpoint.root).state_root`.
4. The state root at slot of `start_slot(n)`, which would be the state from (1), but "skipped forward" through any skip slots.
Previously, Lighthouse would choose (2). However, we can see that when [Teku generates that event](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/beaconrestapi/src/main/java/tech/pegasys/teku/beaconrestapi/handlers/v1/events/EventSubscriptionManager.java#L171-L182) it uses [`getStateRootFromBlockRoot`](https://github.com/ConsenSys/teku/blob/de2b2801c89ef5abf983d6bf37867c37fc47121f/data/provider/src/main/java/tech/pegasys/teku/api/ChainDataProvider.java#L336-L341) which uses (1).
I have switched Lighthouse from (2) to (1). I think it's a somewhat arbitrary choice between the two, where (1) is easier to compute and is consistent with Teku.
## Notes for Reviewers
I've renamed `BeaconChain::fork_choice` to `BeaconChain::recompute_head`. Doing this helped ensure I broke all previous uses of fork choice and I also find it more descriptive. It describes an action and can't be confused with trying to get a reference to the `ForkChoice` struct.
I've changed the ordering of SSE events when a block is received. It used to be `[block, finalized, head]` and now it's `[block, head, finalized]`. It was easier this way and I don't think we were making any promises about SSE event ordering so it's not "breaking".
I've made it so fork choice will run when it's first constructed. I did this because I wanted to have a cached version of the last call to `get_head`. Ensuring `get_head` has been run *at least once* means that the cached values doesn't need to wrapped in an `Option`. This was fairly simple, it just involved passing a `slot` to the constructor so it knows *when* it's being run. When loading a fork choice from the store and a slot clock isn't handy I've just used the `slot` that was saved in the `fork_choice_store`. That seems like it would be a faithful representation of the slot when we saved it.
I added the `genesis_time: u64` to the `BeaconChain`. It's small, constant and nice to have around.
Since we're using FC for the fin/just checkpoints, we no longer get the `0x00..00` roots at genesis. You can see I had to remove a work-around in `ef-tests` here: b56be3bc2. I can't find any reason why this would be an issue, if anything I think it'll be better since the genesis-alias has caught us out a few times (0x00..00 isn't actually a real root). Edit: I did find a case where the `network` expected the 0x00..00 alias and patched it here: 3f26ac3e2.
You'll notice a lot of changes in tests. Generally, tests should be functionally equivalent. Here are the things creating the most diff-noise in tests:
- Changing tests to be `tokio::async` tests.
- Adding `.await` to fork choice, block processing and block production functions.
- Refactor of the `canonical_head` "API" provided by the `BeaconChain`. E.g., `chain.canonical_head.cached_head()` instead of `chain.canonical_head.read()`.
- Wrapping `SignedBeaconBlock` in an `Arc`.
- In the `beacon_chain/tests/block_verification`, we can't use the `lazy_static` `CHAIN_SEGMENT` variable anymore since it's generated with an async function. We just generate it in each test, not so efficient but hopefully insignificant.
I had to disable `rayon` concurrent tests in the `fork_choice` tests. This is because the use of `rayon` and `block_on` was causing a panic.
Co-authored-by: Mac L <mjladson@pm.me>
2022-07-03 05:36:50 +00:00
|
|
|
.await
|
2021-08-06 00:47:31 +00:00
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
let current_period = fork_epoch.sync_committee_period(&spec).unwrap();
|
|
|
|
let next_period_epoch = spec.epochs_per_sync_committee_period * (current_period + 1);
|
|
|
|
assert!(next_period_epoch > fork_epoch);
|
|
|
|
|
|
|
|
for epoch in [
|
|
|
|
None,
|
|
|
|
Some(fork_epoch),
|
|
|
|
Some(fork_epoch + 1),
|
|
|
|
Some(next_period_epoch),
|
|
|
|
Some(next_period_epoch + 1),
|
|
|
|
] {
|
|
|
|
let committee = client
|
|
|
|
.get_beacon_states_sync_committees(StateId::Head, epoch)
|
|
|
|
.await
|
|
|
|
.unwrap()
|
|
|
|
.data;
|
|
|
|
assert_eq!(committee.validators.len(), E::sync_committee_size());
|
|
|
|
|
|
|
|
assert_eq!(
|
|
|
|
committee.validators,
|
|
|
|
flatten(&committee.validator_aggregates)
|
|
|
|
);
|
|
|
|
}
|
|
|
|
}
|
2023-01-20 23:39:59 +00:00
|
|
|
|
|
|
|
/// Assert that an HTTP API error has the given status code and indexed errors for the given indices.
|
|
|
|
fn assert_server_indexed_error(error: eth2::Error, status_code: u16, indices: Vec<usize>) {
|
|
|
|
let eth2::Error::ServerIndexedMessage(IndexedErrorMessage {
|
|
|
|
code,
|
|
|
|
failures,
|
|
|
|
..
|
|
|
|
}) = error else {
|
|
|
|
panic!("wrong error, expected ServerIndexedMessage, got: {error:?}")
|
|
|
|
};
|
|
|
|
assert_eq!(code, status_code);
|
|
|
|
assert_eq!(failures.len(), indices.len());
|
|
|
|
for (index, failure) in indices.into_iter().zip(failures) {
|
|
|
|
assert_eq!(failure.index, index as u64);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
|
|
async fn bls_to_execution_changes_update_all_around_capella_fork() {
|
|
|
|
let validator_count = 128;
|
|
|
|
let fork_epoch = Epoch::new(2);
|
|
|
|
let spec = capella_spec(fork_epoch);
|
|
|
|
let max_bls_to_execution_changes = E::max_bls_to_execution_changes();
|
|
|
|
let tester = InteractiveTester::<E>::new(Some(spec.clone()), validator_count).await;
|
|
|
|
let harness = &tester.harness;
|
|
|
|
let client = &tester.client;
|
|
|
|
|
|
|
|
let all_validators = harness.get_all_validators();
|
|
|
|
let all_validators_u64 = all_validators.iter().map(|x| *x as u64).collect::<Vec<_>>();
|
|
|
|
|
|
|
|
// Create a bunch of valid address changes.
|
|
|
|
let valid_address_changes = all_validators_u64
|
|
|
|
.iter()
|
|
|
|
.map(|&validator_index| {
|
|
|
|
harness.make_bls_to_execution_change(
|
|
|
|
validator_index,
|
|
|
|
Address::from_low_u64_be(validator_index),
|
|
|
|
)
|
|
|
|
})
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
|
|
|
|
// Address changes which conflict with `valid_address_changes` on the address chosen.
|
|
|
|
let conflicting_address_changes = all_validators_u64
|
|
|
|
.iter()
|
|
|
|
.map(|&validator_index| {
|
|
|
|
harness.make_bls_to_execution_change(
|
|
|
|
validator_index,
|
|
|
|
Address::from_low_u64_be(validator_index + 1),
|
|
|
|
)
|
|
|
|
})
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
|
|
|
|
// Address changes signed with the wrong key.
|
|
|
|
let wrong_key_address_changes = all_validators_u64
|
|
|
|
.iter()
|
|
|
|
.map(|&validator_index| {
|
|
|
|
// Use the correct pubkey.
|
|
|
|
let pubkey = &harness.get_withdrawal_keypair(validator_index).pk;
|
|
|
|
// And the wrong secret key.
|
|
|
|
let secret_key = &harness
|
|
|
|
.get_withdrawal_keypair((validator_index + 1) % validator_count as u64)
|
|
|
|
.sk;
|
|
|
|
harness.make_bls_to_execution_change_with_keys(
|
|
|
|
validator_index,
|
|
|
|
Address::from_low_u64_be(validator_index),
|
|
|
|
pubkey,
|
|
|
|
secret_key,
|
|
|
|
)
|
|
|
|
})
|
|
|
|
.collect::<Vec<_>>();
|
|
|
|
|
|
|
|
// Submit some changes before Capella. Just enough to fill two blocks.
|
|
|
|
let num_pre_capella = validator_count / 4;
|
|
|
|
let blocks_filled_pre_capella = 2;
|
|
|
|
assert_eq!(
|
|
|
|
num_pre_capella,
|
|
|
|
blocks_filled_pre_capella * max_bls_to_execution_changes
|
|
|
|
);
|
|
|
|
|
|
|
|
client
|
|
|
|
.post_beacon_pool_bls_to_execution_changes(&valid_address_changes[..num_pre_capella])
|
|
|
|
.await
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
// Conflicting changes for the same validators should all fail.
|
|
|
|
let error = client
|
|
|
|
.post_beacon_pool_bls_to_execution_changes(&conflicting_address_changes[..num_pre_capella])
|
|
|
|
.await
|
|
|
|
.unwrap_err();
|
|
|
|
assert_server_indexed_error(error, 400, (0..num_pre_capella).collect());
|
|
|
|
|
|
|
|
// Re-submitting the same changes should be accepted.
|
|
|
|
client
|
|
|
|
.post_beacon_pool_bls_to_execution_changes(&valid_address_changes[..num_pre_capella])
|
|
|
|
.await
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
// Invalid changes signed with the wrong keys should all be rejected without affecting the seen
|
|
|
|
// indices filters (apply ALL of them).
|
|
|
|
let error = client
|
|
|
|
.post_beacon_pool_bls_to_execution_changes(&wrong_key_address_changes)
|
|
|
|
.await
|
|
|
|
.unwrap_err();
|
|
|
|
assert_server_indexed_error(error, 400, all_validators.clone());
|
|
|
|
|
|
|
|
// Advance to right before Capella.
|
|
|
|
let capella_slot = fork_epoch.start_slot(E::slots_per_epoch());
|
|
|
|
harness.extend_to_slot(capella_slot - 1).await;
|
|
|
|
assert_eq!(harness.head_slot(), capella_slot - 1);
|
|
|
|
|
|
|
|
// Add Capella blocks which should be full of BLS to execution changes.
|
|
|
|
for i in 0..validator_count / max_bls_to_execution_changes {
|
|
|
|
let head_block_root = harness.extend_slots(1).await;
|
|
|
|
let head_block = harness
|
|
|
|
.chain
|
|
|
|
.get_block(&head_block_root)
|
|
|
|
.await
|
|
|
|
.unwrap()
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
let bls_to_execution_changes = head_block
|
|
|
|
.message()
|
|
|
|
.body()
|
|
|
|
.bls_to_execution_changes()
|
|
|
|
.unwrap();
|
|
|
|
|
|
|
|
// Block should be full.
|
|
|
|
assert_eq!(
|
|
|
|
bls_to_execution_changes.len(),
|
|
|
|
max_bls_to_execution_changes,
|
|
|
|
"block not full on iteration {i}"
|
|
|
|
);
|
|
|
|
|
|
|
|
// Included changes should be the ones from `valid_address_changes` in any order.
|
|
|
|
for address_change in bls_to_execution_changes.iter() {
|
|
|
|
assert!(valid_address_changes.contains(address_change));
|
|
|
|
}
|
|
|
|
|
|
|
|
// After the initial 2 blocks, add the rest of the changes using a large
|
|
|
|
// request containing all the valid, all the conflicting and all the invalid.
|
|
|
|
// Despite the invalid and duplicate messages, the new ones should still get picked up by
|
|
|
|
// the pool.
|
|
|
|
if i == blocks_filled_pre_capella - 1 {
|
|
|
|
let all_address_changes: Vec<_> = [
|
|
|
|
valid_address_changes.clone(),
|
|
|
|
conflicting_address_changes.clone(),
|
|
|
|
wrong_key_address_changes.clone(),
|
|
|
|
]
|
|
|
|
.concat();
|
|
|
|
|
|
|
|
let error = client
|
|
|
|
.post_beacon_pool_bls_to_execution_changes(&all_address_changes)
|
|
|
|
.await
|
|
|
|
.unwrap_err();
|
|
|
|
assert_server_indexed_error(
|
|
|
|
error,
|
|
|
|
400,
|
|
|
|
(validator_count..3 * validator_count).collect(),
|
|
|
|
);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Eventually all validators should have eth1 withdrawal credentials.
|
|
|
|
let head_state = harness.get_current_state();
|
|
|
|
for validator in head_state.validators() {
|
|
|
|
assert!(validator.has_eth1_withdrawal_credential(&spec));
|
|
|
|
}
|
|
|
|
}
|