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 😅)
- 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](de2b2801c8/data/beaconrestapi/src/main/java/tech/pegasys/teku/beaconrestapi/handlers/v1/events/EventSubscriptionManager.java (L171-L182)) it uses [`getStateRootFromBlockRoot`](de2b2801c8/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>
This commit is contained in:
Paul Hauner
2022-07-03 05:36:50 +00:00
parent e5212f1320
commit be4e261e74
106 changed files with 6515 additions and 4538 deletions

View File

@@ -81,11 +81,23 @@ pub struct Cases<T> {
}
impl<T: Case> Cases<T> {
pub fn test_results(&self, fork_name: ForkName) -> Vec<CaseResult> {
self.test_cases
.into_par_iter()
.enumerate()
.map(|(i, (ref path, ref tc))| CaseResult::new(i, path, tc, tc.result(i, fork_name)))
.collect()
pub fn test_results(&self, fork_name: ForkName, use_rayon: bool) -> Vec<CaseResult> {
if use_rayon {
self.test_cases
.into_par_iter()
.enumerate()
.map(|(i, (ref path, ref tc))| {
CaseResult::new(i, path, tc, tc.result(i, fork_name))
})
.collect()
} else {
self.test_cases
.iter()
.enumerate()
.map(|(i, (ref path, ref tc))| {
CaseResult::new(i, path, tc, tc.result(i, fork_name))
})
.collect()
}
}
}

View File

@@ -7,15 +7,17 @@ use beacon_chain::{
obtain_indexed_attestation_and_committees_per_slot, VerifiedAttestation,
},
test_utils::{BeaconChainHarness, EphemeralHarnessType},
BeaconChainTypes, HeadInfo,
BeaconChainTypes, CachedHead,
};
use serde_derive::Deserialize;
use ssz_derive::Decode;
use state_processing::state_advance::complete_state_advance;
use std::future::Future;
use std::sync::Arc;
use std::time::Duration;
use types::{
Attestation, BeaconBlock, BeaconState, Checkpoint, Epoch, EthSpec, ExecutionBlockHash,
ForkName, Hash256, IndexedAttestation, SignedBeaconBlock, Slot, Uint256,
Attestation, BeaconBlock, BeaconState, Checkpoint, EthSpec, ExecutionBlockHash, ForkName,
Hash256, IndexedAttestation, SignedBeaconBlock, Slot, Uint256,
};
#[derive(Default, Debug, PartialEq, Clone, Deserialize, Decode)]
@@ -287,19 +289,20 @@ impl<E: EthSpec> Tester<E> {
Ok(self.spec.genesis_slot + slots_since_genesis)
}
fn find_head(&self) -> Result<HeadInfo, Error> {
fn block_on_dangerous<F: Future>(&self, future: F) -> Result<F::Output, Error> {
self.harness
.chain
.fork_choice()
.map_err(|e| Error::InternalError(format!("failed to find head with {:?}", e)))?;
self.harness
.chain
.head_info()
.map_err(|e| Error::InternalError(format!("failed to read head with {:?}", e)))
.task_executor
.clone()
.block_on_dangerous(future, "ef_tests_block_on")
.ok_or_else(|| Error::InternalError("runtime shutdown".into()))
}
fn genesis_epoch(&self) -> Epoch {
self.spec.genesis_slot.epoch(E::slots_per_epoch())
fn find_head(&self) -> Result<CachedHead<E>, Error> {
let chain = self.harness.chain.clone();
self.block_on_dangerous(chain.recompute_head_at_current_slot())?
.map_err(|e| Error::InternalError(format!("failed to find head with {:?}", e)))?;
Ok(self.harness.chain.canonical_head.cached_head())
}
pub fn set_tick(&self, tick: u64) {
@@ -314,15 +317,16 @@ impl<E: EthSpec> Tester<E> {
self.harness
.chain
.fork_choice
.write()
.canonical_head
.fork_choice_write_lock()
.update_time(slot)
.unwrap();
}
pub fn process_block(&self, block: SignedBeaconBlock<E>, valid: bool) -> Result<(), Error> {
let result = self.harness.chain.process_block(block.clone());
let block_root = block.canonical_root();
let block = Arc::new(block);
let result = self.block_on_dangerous(self.harness.chain.process_block(block.clone()))?;
if result.is_ok() != valid {
return Err(Error::DidntFail(format!(
"block with root {} was valid={} whilst test expects valid={}. result: {:?}",
@@ -367,16 +371,20 @@ impl<E: EthSpec> Tester<E> {
.seconds_from_current_slot_start(self.spec.seconds_per_slot)
.unwrap();
let (block, _) = block.deconstruct();
let result = self.harness.chain.fork_choice.write().on_block(
self.harness.chain.slot().unwrap(),
&block,
block_root,
block_delay,
&state,
PayloadVerificationStatus::Irrelevant,
&self.harness.chain.spec,
);
let result = self
.harness
.chain
.canonical_head
.fork_choice_write_lock()
.on_block(
self.harness.chain.slot().unwrap(),
block.message(),
block_root,
block_delay,
&state,
PayloadVerificationStatus::Irrelevant,
&self.harness.chain.spec,
);
if result.is_ok() {
return Err(Error::DidntFail(format!(
@@ -424,10 +432,11 @@ impl<E: EthSpec> Tester<E> {
}
pub fn check_head(&self, expected_head: Head) -> Result<(), Error> {
let chain_head = self.find_head().map(|head| Head {
slot: head.slot,
root: head.block_root,
})?;
let head = self.find_head()?;
let chain_head = Head {
slot: head.head_slot(),
root: head.head_block_root(),
};
check_equal("head", chain_head, expected_head)
}
@@ -446,15 +455,15 @@ impl<E: EthSpec> Tester<E> {
}
pub fn check_justified_checkpoint(&self, expected_checkpoint: Checkpoint) -> Result<(), Error> {
let head_checkpoint = self.find_head()?.current_justified_checkpoint;
let fc_checkpoint = self.harness.chain.fork_choice.read().justified_checkpoint();
let head_checkpoint = self.find_head()?.justified_checkpoint();
let fc_checkpoint = self
.harness
.chain
.canonical_head
.fork_choice_read_lock()
.justified_checkpoint();
assert_checkpoints_eq(
"justified_checkpoint",
self.genesis_epoch(),
head_checkpoint,
fc_checkpoint,
);
assert_checkpoints_eq("justified_checkpoint", head_checkpoint, fc_checkpoint);
check_equal("justified_checkpoint", fc_checkpoint, expected_checkpoint)
}
@@ -463,15 +472,15 @@ impl<E: EthSpec> Tester<E> {
&self,
expected_checkpoint_root: Hash256,
) -> Result<(), Error> {
let head_checkpoint = self.find_head()?.current_justified_checkpoint;
let fc_checkpoint = self.harness.chain.fork_choice.read().justified_checkpoint();
let head_checkpoint = self.find_head()?.justified_checkpoint();
let fc_checkpoint = self
.harness
.chain
.canonical_head
.fork_choice_read_lock()
.justified_checkpoint();
assert_checkpoints_eq(
"justified_checkpoint_root",
self.genesis_epoch(),
head_checkpoint,
fc_checkpoint,
);
assert_checkpoints_eq("justified_checkpoint_root", head_checkpoint, fc_checkpoint);
check_equal(
"justified_checkpoint_root",
@@ -481,15 +490,15 @@ impl<E: EthSpec> Tester<E> {
}
pub fn check_finalized_checkpoint(&self, expected_checkpoint: Checkpoint) -> Result<(), Error> {
let head_checkpoint = self.find_head()?.finalized_checkpoint;
let fc_checkpoint = self.harness.chain.fork_choice.read().finalized_checkpoint();
let head_checkpoint = self.find_head()?.finalized_checkpoint();
let fc_checkpoint = self
.harness
.chain
.canonical_head
.fork_choice_read_lock()
.finalized_checkpoint();
assert_checkpoints_eq(
"finalized_checkpoint",
self.genesis_epoch(),
head_checkpoint,
fc_checkpoint,
);
assert_checkpoints_eq("finalized_checkpoint", head_checkpoint, fc_checkpoint);
check_equal("finalized_checkpoint", fc_checkpoint, expected_checkpoint)
}
@@ -501,8 +510,8 @@ impl<E: EthSpec> Tester<E> {
let best_justified_checkpoint = self
.harness
.chain
.fork_choice
.read()
.canonical_head
.fork_choice_read_lock()
.best_justified_checkpoint();
check_equal(
"best_justified_checkpoint",
@@ -515,7 +524,12 @@ impl<E: EthSpec> Tester<E> {
&self,
expected_proposer_boost_root: Hash256,
) -> Result<(), Error> {
let proposer_boost_root = self.harness.chain.fork_choice.read().proposer_boost_root();
let proposer_boost_root = self
.harness
.chain
.canonical_head
.fork_choice_read_lock()
.proposer_boost_root();
check_equal(
"proposer_boost_root",
proposer_boost_root,
@@ -530,20 +544,8 @@ impl<E: EthSpec> Tester<E> {
/// This function is necessary due to a quirk documented in this issue:
///
/// https://github.com/ethereum/consensus-specs/issues/2566
fn assert_checkpoints_eq(name: &str, genesis_epoch: Epoch, head: Checkpoint, fc: Checkpoint) {
if fc.epoch == genesis_epoch {
assert_eq!(
head,
Checkpoint {
epoch: genesis_epoch,
root: Hash256::zero()
},
"{} (genesis)",
name
)
} else {
assert_eq!(head, fc, "{} (non-genesis)", name)
}
fn assert_checkpoints_eq(name: &str, head: Checkpoint, fc: Checkpoint) {
assert_eq!(head, fc, "{}", name)
}
/// Convenience function to create `Error` messages.

View File

@@ -30,6 +30,10 @@ pub trait Handler {
}
}
fn use_rayon() -> bool {
true
}
fn run_for_fork(&self, fork_name: ForkName) {
let fork_name_str = fork_name.to_string();
@@ -59,7 +63,7 @@ pub trait Handler {
})
.collect();
let results = Cases { test_cases }.test_results(fork_name);
let results = Cases { test_cases }.test_results(fork_name, Self::use_rayon());
let name = format!(
"{}/{}/{}",
@@ -460,6 +464,11 @@ impl<E: EthSpec + TypeName> Handler for ForkChoiceHandler<E> {
self.handler_name.clone()
}
fn use_rayon() -> bool {
// The fork choice tests use `block_on` which can cause panics with rayon.
false
}
fn is_enabled_for_fork(&self, fork_name: ForkName) -> bool {
// Merge block tests are only enabled for Bellatrix or later.
if self.handler_name == "on_merge_block"

View File

@@ -98,10 +98,9 @@ impl<E: GenericExecutionEngine> TestRig<E> {
}
pub fn perform_tests_blocking(&self) {
self.ee_a
.execution_layer
.block_on_generic(|_| async { self.perform_tests().await })
.unwrap()
self.runtime
.handle()
.block_on(async { self.perform_tests().await });
}
pub async fn wait_until_synced(&self) {

View File

@@ -12,3 +12,4 @@ types = { path = "../../consensus/types" }
eth2_ssz = "0.4.1"
beacon_chain = { path = "../../beacon_node/beacon_chain" }
lazy_static = "1.4.0"
tokio = { version = "1.14.0", features = ["rt-multi-thread"] }

View File

@@ -37,11 +37,12 @@ impl Default for ExitTest {
}
impl ExitTest {
fn block_and_pre_state(self) -> (SignedBeaconBlock<E>, BeaconState<E>) {
async fn block_and_pre_state(self) -> (SignedBeaconBlock<E>, BeaconState<E>) {
let harness = get_harness::<E>(
self.state_epoch.start_slot(E::slots_per_epoch()),
VALIDATOR_COUNT,
);
)
.await;
let mut state = harness.get_current_state();
(self.state_modifier)(&mut state);
@@ -49,11 +50,12 @@ impl ExitTest {
let validator_index = self.validator_index;
let exit_epoch = self.exit_epoch;
let (signed_block, state) =
harness.make_block_with_modifier(state.clone(), state.slot() + 1, |block| {
let (signed_block, state) = harness
.make_block_with_modifier(state.clone(), state.slot() + 1, |block| {
harness.add_voluntary_exit(block, validator_index, exit_epoch);
block_modifier(&harness, block);
});
})
.await;
(signed_block, state)
}
@@ -72,12 +74,12 @@ impl ExitTest {
}
#[cfg(all(test, not(debug_assertions)))]
fn run(self) -> BeaconState<E> {
async fn run(self) -> BeaconState<E> {
let spec = &E::default_spec();
let expected = self.expected.clone();
assert_eq!(STATE_EPOCH, spec.shard_committee_period);
let (block, mut state) = self.block_and_pre_state();
let (block, mut state) = self.block_and_pre_state().await;
let result = Self::process(&block, &mut state);
@@ -86,8 +88,8 @@ impl ExitTest {
state
}
fn test_vector(self, title: String) -> TestVector {
let (block, pre_state) = self.block_and_pre_state();
async fn test_vector(self, title: String) -> TestVector {
let (block, pre_state) = self.block_and_pre_state().await;
let mut post_state = pre_state.clone();
let (post_state, error) = match Self::process(&block, &mut post_state) {
Ok(_) => (Some(post_state), None),
@@ -334,14 +336,14 @@ mod custom_tests {
);
}
#[test]
fn valid() {
let state = ExitTest::default().run();
#[tokio::test]
async fn valid() {
let state = ExitTest::default().run().await;
assert_exited(&state, VALIDATOR_INDEX as usize);
}
#[test]
fn valid_three() {
#[tokio::test]
async fn valid_three() {
let state = ExitTest {
block_modifier: Box::new(|harness, block| {
harness.add_voluntary_exit(block, 1, STATE_EPOCH);
@@ -349,7 +351,8 @@ mod custom_tests {
}),
..ExitTest::default()
}
.run();
.run()
.await;
for i in &[VALIDATOR_INDEX, 1, 2] {
assert_exited(&state, *i as usize);

View File

@@ -4,11 +4,11 @@
/// - `mod tests`: runs all the test vectors locally.
macro_rules! vectors_and_tests {
($($name: ident, $test: expr),*) => {
pub fn vectors() -> Vec<TestVector> {
pub async fn vectors() -> Vec<TestVector> {
let mut vec = vec![];
$(
vec.push($test.test_vector(stringify!($name).into()));
vec.push($test.test_vector(stringify!($name).into()).await);
)*
vec
@@ -18,9 +18,9 @@ macro_rules! vectors_and_tests {
mod tests {
use super::*;
$(
#[test]
fn $name() {
$test.run();
#[tokio::test]
async fn $name() {
$test.run().await;
}
)*
}

View File

@@ -25,8 +25,9 @@ pub const BASE_VECTOR_DIR: &str = "vectors";
pub const SLOT_OFFSET: u64 = 1;
/// Writes all known test vectors to `CARGO_MANIFEST_DIR/vectors`.
fn main() {
match write_all_vectors() {
#[tokio::main]
async fn main() {
match write_all_vectors().await {
Ok(()) => exit(0),
Err(e) => {
eprintln!("Error: {}", e);
@@ -49,7 +50,7 @@ lazy_static! {
static ref KEYPAIRS: Vec<Keypair> = generate_deterministic_keypairs(VALIDATOR_COUNT);
}
fn get_harness<E: EthSpec>(
async fn get_harness<E: EthSpec>(
slot: Slot,
validator_count: usize,
) -> BeaconChainHarness<EphemeralHarnessType<E>> {
@@ -61,23 +62,25 @@ fn get_harness<E: EthSpec>(
let skip_to_slot = slot - SLOT_OFFSET;
if skip_to_slot > Slot::new(0) {
let state = harness.get_current_state();
harness.add_attested_blocks_at_slots(
state,
Hash256::zero(),
(skip_to_slot.as_u64()..slot.as_u64())
.map(Slot::new)
.collect::<Vec<_>>()
.as_slice(),
(0..validator_count).collect::<Vec<_>>().as_slice(),
);
harness
.add_attested_blocks_at_slots(
state,
Hash256::zero(),
(skip_to_slot.as_u64()..slot.as_u64())
.map(Slot::new)
.collect::<Vec<_>>()
.as_slice(),
(0..validator_count).collect::<Vec<_>>().as_slice(),
)
.await;
}
harness
}
/// Writes all vectors to file.
fn write_all_vectors() -> Result<(), String> {
write_vectors_to_file("exit", &exit::vectors())
async fn write_all_vectors() -> Result<(), String> {
write_vectors_to_file("exit", &exit::vectors().await)
}
/// Writes a list of `vectors` to the `title` dir.