lib.rs

// Copyright (C) Parity Technologies (UK) Ltd.
// This file is part of Polkadot.

// Polkadot is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.

// Polkadot is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.

// You should have received a copy of the GNU General Public License
// along with Polkadot.  If not, see <http://www.gnu.org/licenses/>.

//! The collation generation subsystem is the interface between polkadot and the collators.
//!
//! # Protocol
//!
//! On every `ActiveLeavesUpdate`:
//!
//! * If there is no collation generation config, ignore.
//! * Otherwise, for each `activated` head in the update:
//!   * Determine if the para is scheduled on any core by fetching the `availability_cores` Runtime API.
//!   * Determine an occupied core assumption to make about the para. Scheduled cores can make [`OccupiedCoreAssumption::Free`].
//!     * TODO: What does this mean?
//!   * Use the Runtime API subsystem to fetch the full validation data.
//!   * Invoke the `collator`, and use its outputs to produce a [`CandidateReceipt`], signed with the configuration's `key`.
//!   * Dispatch a [`CollatorProtocolMessage::DistributeCollation`](receipt, pov)`.

#![deny(missing_docs)]

use futures::{channel::mpsc, future::FutureExt, join, select, sink::SinkExt, stream::StreamExt};
use parity_scale_codec::Encode;
use polkadot_node_primitives::{AvailableData, CollationGenerationConfig, PoV};
use polkadot_node_subsystem::{
	messages::{CollationGenerationMessage, CollatorProtocolMessage},
	overseer, ActiveLeavesUpdate, FromOrchestra, OverseerSignal, SpawnedSubsystem,
	SubsystemContext, SubsystemError, SubsystemResult,
};
use polkadot_node_subsystem_util::{
	request_availability_cores, request_persisted_validation_data, request_validation_code,
	request_validation_code_hash, request_validators,
};
use polkadot_primitives::{
	collator_signature_payload, CandidateCommitments, CandidateDescriptor, CandidateReceipt,
	CoreState, Hash, Id as ParaId, OccupiedCoreAssumption, PersistedValidationData,
	ValidationCodeHash,
};
use sp_core::crypto::Pair;
use std::sync::Arc;

mod error;

#[cfg(test)]
mod tests;

mod metrics;
use self::metrics::Metrics;

const LOG_TARGET: &'static str = "parachain::collation-generation";

/// Collation Generation Subsystem
pub struct CollationGenerationSubsystem {
	config: Option<Arc<CollationGenerationConfig>>,
	metrics: Metrics,
}

#[overseer::contextbounds(CollationGeneration, prefix = self::overseer)]
impl CollationGenerationSubsystem {
	/// Create a new instance of the `CollationGenerationSubsystem`.
	pub fn new(metrics: Metrics) -> Self {
		Self { config: None, metrics }
	}

	/// Run this subsystem
	///
	/// Conceptually, this is very simple: it just loops forever.
	///
	/// - On incoming overseer messages, it starts or stops jobs as appropriate.
	/// - On other incoming messages, if they can be converted into `Job::ToJob` and
	///   include a hash, then they're forwarded to the appropriate individual job.
	/// - On outgoing messages from the jobs, it forwards them to the overseer.
	///
	/// If `err_tx` is not `None`, errors are forwarded onto that channel as they occur.
	/// Otherwise, most are logged and then discarded.
	async fn run<Context>(mut self, mut ctx: Context) {
		// when we activate new leaves, we spawn a bunch of sub-tasks, each of which is
		// expected to generate precisely one message. We don't want to block the main loop
		// at any point waiting for them all, so instead, we create a channel on which they can
		// send those messages. We can then just monitor the channel and forward messages on it
		// to the overseer here, via the context.
		let (sender, receiver) = mpsc::channel(0);

		let mut receiver = receiver.fuse();
		loop {
			select! {
				incoming = ctx.recv().fuse() => {
					if self.handle_incoming::<Context>(incoming, &mut ctx, &sender).await {
						break;
					}
				},
				msg = receiver.next() => {
					if let Some(msg) = msg {
						ctx.send_message(msg).await;
					}
				},
			}
		}
	}

	// handle an incoming message. return true if we should break afterwards.
	// note: this doesn't strictly need to be a separate function; it's more an administrative function
	// so that we don't clutter the run loop. It could in principle be inlined directly into there.
	// it should hopefully therefore be ok that it's an async function mutably borrowing self.
	async fn handle_incoming<Context>(
		&mut self,
		incoming: SubsystemResult<FromOrchestra<<Context as SubsystemContext>::Message>>,
		ctx: &mut Context,
		sender: &mpsc::Sender<overseer::CollationGenerationOutgoingMessages>,
	) -> bool {
		match incoming {
			Ok(FromOrchestra::Signal(OverseerSignal::ActiveLeaves(ActiveLeavesUpdate {
				activated,
				..
			}))) => {
				// follow the procedure from the guide
				if let Some(config) = &self.config {
					let metrics = self.metrics.clone();
					if let Err(err) = handle_new_activations(
						config.clone(),
						activated.into_iter().map(|v| v.hash),
						ctx,
						metrics,
						sender,
					)
					.await
					{
						gum::warn!(target: LOG_TARGET, err = ?err, "failed to handle new activations");
					}
				}

				false
			},
			Ok(FromOrchestra::Signal(OverseerSignal::Conclude)) => true,
			Ok(FromOrchestra::Communication {
				msg: CollationGenerationMessage::Initialize(config),
			}) => {
				if self.config.is_some() {
					gum::error!(target: LOG_TARGET, "double initialization");
				} else {
					self.config = Some(Arc::new(config));
				}
				false
			},
			Ok(FromOrchestra::Signal(OverseerSignal::BlockFinalized(..))) => false,
			Err(err) => {
				gum::error!(
					target: LOG_TARGET,
					err = ?err,
					"error receiving message from subsystem context: {:?}",
					err
				);
				true
			},
		}
	}
}

#[overseer::subsystem(CollationGeneration, error=SubsystemError, prefix=self::overseer)]
impl<Context> CollationGenerationSubsystem {
	fn start(self, ctx: Context) -> SpawnedSubsystem {
		let future = async move {
			self.run(ctx).await;
			Ok(())
		}
		.boxed();

		SpawnedSubsystem { name: "collation-generation-subsystem", future }
	}
}

#[overseer::contextbounds(CollationGeneration, prefix = self::overseer)]
async fn handle_new_activations<Context>(
	config: Arc<CollationGenerationConfig>,
	activated: impl IntoIterator<Item = Hash>,
	ctx: &mut Context,
	metrics: Metrics,
	sender: &mpsc::Sender<overseer::CollationGenerationOutgoingMessages>,
) -> crate::error::Result<()> {
	// follow the procedure from the guide:
	// https://paritytech.github.io/polkadot/book/node/collators/collation-generation.html

	let _overall_timer = metrics.time_new_activations();

	for relay_parent in activated {
		let _relay_parent_timer = metrics.time_new_activations_relay_parent();

		let (availability_cores, validators) = join!(
			request_availability_cores(relay_parent, ctx.sender()).await,
			request_validators(relay_parent, ctx.sender()).await,
		);

		let availability_cores = availability_cores??;
		let n_validators = validators??.len();

		for (core_idx, core) in availability_cores.into_iter().enumerate() {
			let _availability_core_timer = metrics.time_new_activations_availability_core();

			let (scheduled_core, assumption) = match core {
				CoreState::Scheduled(scheduled_core) =>
					(scheduled_core, OccupiedCoreAssumption::Free),
				CoreState::Occupied(_occupied_core) => {
					// TODO: https://github.com/paritytech/polkadot/issues/1573
					gum::trace!(
						target: LOG_TARGET,
						core_idx = %core_idx,
						relay_parent = ?relay_parent,
						"core is occupied. Keep going.",
					);
					continue
				},
				CoreState::Free => {
					gum::trace!(
						target: LOG_TARGET,
						core_idx = %core_idx,
						"core is free. Keep going.",
					);
					continue
				},
			};

			if scheduled_core.para_id != config.para_id {
				gum::trace!(
					target: LOG_TARGET,
					core_idx = %core_idx,
					relay_parent = ?relay_parent,
					our_para = %config.para_id,
					their_para = %scheduled_core.para_id,
					"core is not assigned to our para. Keep going.",
				);
				continue
			}

			// we get validation data and validation code synchronously for each core instead of
			// within the subtask loop, because we have only a single mutable handle to the
			// context, so the work can't really be distributed

			let validation_data = match request_persisted_validation_data(
				relay_parent,
				scheduled_core.para_id,
				assumption,
				ctx.sender(),
			)
			.await
			.await??
			{
				Some(v) => v,
				None => {
					gum::trace!(
						target: LOG_TARGET,
						core_idx = %core_idx,
						relay_parent = ?relay_parent,
						our_para = %config.para_id,
						their_para = %scheduled_core.para_id,
						"validation data is not available",
					);
					continue
				},
			};

			let validation_code_hash = match obtain_current_validation_code_hash(
				relay_parent,
				scheduled_core.para_id,
				assumption,
				ctx.sender(),
			)
			.await?
			{
				Some(v) => v,
				None => {
					gum::trace!(
						target: LOG_TARGET,
						core_idx = %core_idx,
						relay_parent = ?relay_parent,
						our_para = %config.para_id,
						their_para = %scheduled_core.para_id,
						"validation code hash is not found.",
					);
					continue
				},
			};

			let task_config = config.clone();
			let mut task_sender = sender.clone();
			let metrics = metrics.clone();
			ctx.spawn(
				"collation-builder",
				Box::pin(async move {
					let persisted_validation_data_hash = validation_data.hash();

					let (collation, result_sender) =
						match (task_config.collator)(relay_parent, &validation_data).await {
							Some(collation) => collation.into_inner(),
							None => {
								gum::debug!(
									target: LOG_TARGET,
									para_id = %scheduled_core.para_id,
									"collator returned no collation on collate",
								);
								return
							},
						};

					// Apply compression to the block data.
					let pov = {
						let pov = collation.proof_of_validity.into_compressed();
						let encoded_size = pov.encoded_size();

						// As long as `POV_BOMB_LIMIT` is at least `max_pov_size`, this ensures
						// that honest collators never produce a PoV which is uncompressed.
						//
						// As such, honest collators never produce an uncompressed PoV which starts with
						// a compression magic number, which would lead validators to reject the collation.
						if encoded_size > validation_data.max_pov_size as usize {
							gum::debug!(
								target: LOG_TARGET,
								para_id = %scheduled_core.para_id,
								size = encoded_size,
								max_size = validation_data.max_pov_size,
								"PoV exceeded maximum size"
							);

							return
						}

						pov
					};

					let pov_hash = pov.hash();

					let signature_payload = collator_signature_payload(
						&relay_parent,
						&scheduled_core.para_id,
						&persisted_validation_data_hash,
						&pov_hash,
						&validation_code_hash,
					);

					let erasure_root =
						match erasure_root(n_validators, validation_data, pov.clone()) {
							Ok(erasure_root) => erasure_root,
							Err(err) => {
								gum::error!(
									target: LOG_TARGET,
									para_id = %scheduled_core.para_id,
									err = ?err,
									"failed to calculate erasure root",
								);
								return
							},
						};

					let commitments = CandidateCommitments {
						upward_messages: collation.upward_messages,
						horizontal_messages: collation.horizontal_messages,
						new_validation_code: collation.new_validation_code,
						head_data: collation.head_data,
						processed_downward_messages: collation.processed_downward_messages,
						hrmp_watermark: collation.hrmp_watermark,
					};

					let ccr = CandidateReceipt {
						commitments_hash: commitments.hash(),
						descriptor: CandidateDescriptor {
							signature: task_config.key.sign(&signature_payload),
							para_id: scheduled_core.para_id,
							relay_parent,
							collator: task_config.key.public(),
							persisted_validation_data_hash,
							pov_hash,
							erasure_root,
							para_head: commitments.head_data.hash(),
							validation_code_hash,
						},
					};

					gum::debug!(
						target: LOG_TARGET,
						candidate_hash = ?ccr.hash(),
						?pov_hash,
						?relay_parent,
						para_id = %scheduled_core.para_id,
						"candidate is generated",
					);
					metrics.on_collation_generated();

					if let Err(err) = task_sender
						.send(
							CollatorProtocolMessage::DistributeCollation(ccr, pov, result_sender)
								.into(),
						)
						.await
					{
						gum::warn!(
							target: LOG_TARGET,
							para_id = %scheduled_core.para_id,
							err = ?err,
							"failed to send collation result",
						);
					}
				}),
			)?;
		}
	}

	Ok(())
}

async fn obtain_current_validation_code_hash(
	relay_parent: Hash,
	para_id: ParaId,
	assumption: OccupiedCoreAssumption,
	sender: &mut impl overseer::CollationGenerationSenderTrait,
) -> Result<Option<ValidationCodeHash>, crate::error::Error> {
	use polkadot_node_subsystem::RuntimeApiError;

	match request_validation_code_hash(relay_parent, para_id, assumption, sender)
		.await
		.await?
	{
		Ok(Some(v)) => Ok(Some(v)),
		Ok(None) => Ok(None),
		Err(RuntimeApiError::NotSupported { .. }) => {
			match request_validation_code(relay_parent, para_id, assumption, sender).await.await? {
				Ok(Some(v)) => Ok(Some(v.hash())),
				Ok(None) => Ok(None),
				Err(e) => {
					// We assume that the `validation_code` API is always available, so any error
					// is unexpected.
					Err(e.into())
				},
			}
		},
		Err(e @ RuntimeApiError::Execution { .. }) => Err(e.into()),
	}
}

fn erasure_root(
	n_validators: usize,
	persisted_validation: PersistedValidationData,
	pov: PoV,
) -> crate::error::Result<Hash> {
	let available_data =
		AvailableData { validation_data: persisted_validation, pov: Arc::new(pov) };

	let chunks = polkadot_erasure_coding::obtain_chunks_v1(n_validators, &available_data)?;
	Ok(polkadot_erasure_coding::branches(&chunks).root())
}