Integrate litep2p into Polkadot SDK (#2944)

[litep2p](https://github.com/altonen/litep2p) is a libp2p-compatible P2P networking library. It supports all of the features of `rust-libp2p` that are currently being utilized by Polkadot SDK. Compared to `rust-libp2p`, `litep2p` has a quite different architecture which is why the new `litep2p` network backend is only able to use a little of the existing code in `sc-network`. The design has been mainly influenced by how we'd wish to structure our networking-related code in Polkadot SDK: independent higher-levels protocols directly communicating with the network over links that support bidirectional backpressure. A good example would be `NotificationHandle`/`RequestResponseHandle` abstractions which allow, e.g., `SyncingEngine` to directly communicate with peers to announce/request blocks. I've tried running `polkadot --network-backend litep2p` with a few different peer configurations and there is a noticeable reduction in networking CPU usage. For high load (`--out-peers 200`), networking CPU usage goes down from ~110% to ~30% (80 pp) and for normal load (`--out-peers 40`), the usage goes down from ~55% to ~18% (37 pp). These should not be taken as final numbers because: a) there are still some low-hanging optimization fruits, such as enabling [receive window auto-tuning](https://github.com/libp2p/rust-yamux/pull/176), integrating `Peerset` more closely with `litep2p` or improving memory usage of the WebSocket transport b) fixing bugs/instabilities that incorrectly cause `litep2p` to do less work will increase the networking CPU usage c) verification in a more diverse set of tests/conditions is needed Nevertheless, these numbers should give an early estimate for CPU usage of the new networking backend. This PR consists of three separate changes: * introduce a generic `PeerId` (wrapper around `Multihash`) so that we don't have use `NetworkService::PeerId` in every part of the code that uses a `PeerId` * introduce `NetworkBackend` trait, implement it for the libp2p network stack and make Polkadot SDK generic over `NetworkBackend` * implement `NetworkBackend` for litep2p The new library should be considered experimental which is why `rust-libp2p` will remain as the default option for the time being. This PR currently depends on the master branch of `litep2p` but I'll cut a new release for the library once all review comments have been addresses. --------- Signed-off-by: Alexandru Vasile <alexandru.vasile@parity.io> Co-authored-by: Dmitry Markin <dmitry@markin.tech> Co-authored-by: Alexandru Vasile <60601340+lexnv@users.noreply.github.com> Co-authored-by: Alexandru Vasile <alexandru.vasile@parity.io>
2026-04-27 05:47:58 +00:00 · 2024-04-08 19:44:13 +03:00
parent 9543d31474
commit 80616f6d03
181 changed files with 11055 additions and 1862 deletions
@@ -16,7 +16,7 @@

 //! Error handling related code and Error/Result definitions.

-use sc_network::PeerId;
+use sc_network_types::PeerId;

 use parity_scale_codec::Error as DecodingError;

@@ -20,7 +20,9 @@ use futures::{channel::oneshot, StreamExt};

 use parity_scale_codec::{Decode, Encode};

-use sc_network::{config as netconfig, config::RequestResponseConfig, PeerId};
+use sc_network::{config as netconfig, NetworkBackend};
+use sc_network_types::PeerId;
+use sp_runtime::traits::Block;

 use super::{IsRequest, ReqProtocolNames};
 use crate::UnifiedReputationChange;
@@ -52,10 +54,10 @@ where
 	///
 	/// This Register that config with substrate networking and receive incoming requests via the
 	/// returned `IncomingRequestReceiver`.
-	pub fn get_config_receiver(
+	pub fn get_config_receiver<B: Block, N: NetworkBackend<B, <B as Block>::Hash>>(
 		req_protocol_names: &ReqProtocolNames,
-	) -> (IncomingRequestReceiver<Req>, RequestResponseConfig) {
-		let (raw, cfg) = Req::PROTOCOL.get_config(req_protocol_names);
+	) -> (IncomingRequestReceiver<Req>, N::RequestResponseProtocolConfig) {
+		let (raw, cfg) = Req::PROTOCOL.get_config::<B, N>(req_protocol_names);
 		(IncomingRequestReceiver { raw, phantom: PhantomData {} }, cfg)
 	}

@@ -52,6 +52,8 @@
 use std::{collections::HashMap, time::Duration, u64};

 use polkadot_primitives::{MAX_CODE_SIZE, MAX_POV_SIZE};
+use sc_network::NetworkBackend;
+use sp_runtime::traits::Block;
 use strum::{EnumIter, IntoEnumIterator};

 pub use sc_network::{config as network, config::RequestResponseConfig, ProtocolName};
@@ -179,76 +181,76 @@ impl Protocol {
 	///
 	/// Returns a `ProtocolConfig` for this protocol.
 	/// Use this if you plan only to send requests for this protocol.
-	pub fn get_outbound_only_config(
+	pub fn get_outbound_only_config<B: Block, N: NetworkBackend<B, <B as Block>::Hash>>(
 		self,
 		req_protocol_names: &ReqProtocolNames,
-	) -> RequestResponseConfig {
-		self.create_config(req_protocol_names, None)
+	) -> N::RequestResponseProtocolConfig {
+		self.create_config::<B, N>(req_protocol_names, None)
 	}

 	/// Get a configuration for a given Request response protocol.
 	///
 	/// Returns a receiver for messages received on this protocol and the requested
 	/// `ProtocolConfig`.
-	pub fn get_config(
+	pub fn get_config<B: Block, N: NetworkBackend<B, <B as Block>::Hash>>(
 		self,
 		req_protocol_names: &ReqProtocolNames,
-	) -> (async_channel::Receiver<network::IncomingRequest>, RequestResponseConfig) {
+	) -> (async_channel::Receiver<network::IncomingRequest>, N::RequestResponseProtocolConfig) {
 		let (tx, rx) = async_channel::bounded(self.get_channel_size());
-		let cfg = self.create_config(req_protocol_names, Some(tx));
+		let cfg = self.create_config::<B, N>(req_protocol_names, Some(tx));
 		(rx, cfg)
 	}

-	fn create_config(
+	fn create_config<B: Block, N: NetworkBackend<B, <B as Block>::Hash>>(
 		self,
 		req_protocol_names: &ReqProtocolNames,
 		tx: Option<async_channel::Sender<network::IncomingRequest>>,
-	) -> RequestResponseConfig {
+	) -> N::RequestResponseProtocolConfig {
 		let name = req_protocol_names.get_name(self);
 		let legacy_names = self.get_legacy_name().into_iter().map(Into::into).collect();
 		match self {
-			Protocol::ChunkFetchingV1 => RequestResponseConfig {
+			Protocol::ChunkFetchingV1 => N::request_response_config(
 				name,
-				fallback_names: legacy_names,
-				max_request_size: 1_000,
-				max_response_size: POV_RESPONSE_SIZE as u64 * 3,
+				legacy_names,
+				1_000,
+				POV_RESPONSE_SIZE as u64 * 3,
 				// We are connected to all validators:
-				request_timeout: CHUNK_REQUEST_TIMEOUT,
-				inbound_queue: tx,
-			},
+				CHUNK_REQUEST_TIMEOUT,
+				tx,
+			),
 			Protocol::CollationFetchingV1 | Protocol::CollationFetchingV2 =>
-				RequestResponseConfig {
+				N::request_response_config(
 					name,
-					fallback_names: legacy_names,
-					max_request_size: 1_000,
-					max_response_size: POV_RESPONSE_SIZE,
+					legacy_names,
+					1_000,
+					POV_RESPONSE_SIZE,
 					// Taken from initial implementation in collator protocol:
-					request_timeout: POV_REQUEST_TIMEOUT_CONNECTED,
-					inbound_queue: tx,
-				},
-			Protocol::PoVFetchingV1 => RequestResponseConfig {
+					POV_REQUEST_TIMEOUT_CONNECTED,
+					tx,
+				),
+			Protocol::PoVFetchingV1 => N::request_response_config(
 				name,
-				fallback_names: legacy_names,
-				max_request_size: 1_000,
-				max_response_size: POV_RESPONSE_SIZE,
-				request_timeout: POV_REQUEST_TIMEOUT_CONNECTED,
-				inbound_queue: tx,
-			},
-			Protocol::AvailableDataFetchingV1 => RequestResponseConfig {
+				legacy_names,
+				1_000,
+				POV_RESPONSE_SIZE,
+				POV_REQUEST_TIMEOUT_CONNECTED,
+				tx,
+			),
+			Protocol::AvailableDataFetchingV1 => N::request_response_config(
 				name,
-				fallback_names: legacy_names,
-				max_request_size: 1_000,
+				legacy_names,
+				1_000,
 				// Available data size is dominated by the PoV size.
-				max_response_size: POV_RESPONSE_SIZE,
-				request_timeout: POV_REQUEST_TIMEOUT_CONNECTED,
-				inbound_queue: tx,
-			},
-			Protocol::StatementFetchingV1 => RequestResponseConfig {
+				POV_RESPONSE_SIZE,
+				POV_REQUEST_TIMEOUT_CONNECTED,
+				tx,
+			),
+			Protocol::StatementFetchingV1 => N::request_response_config(
 				name,
-				fallback_names: legacy_names,
-				max_request_size: 1_000,
+				legacy_names,
+				1_000,
 				// Available data size is dominated code size.
-				max_response_size: STATEMENT_RESPONSE_SIZE,
+				STATEMENT_RESPONSE_SIZE,
 				// We need statement fetching to be fast and will try our best at the responding
 				// side to answer requests within that timeout, assuming a bandwidth of 500Mbit/s
 				// - which is the recommended minimum bandwidth for nodes on Kusama as of April
@@ -258,27 +260,27 @@ impl Protocol {
 				// waiting for timeout on an overloaded node.  Fetches from slow nodes will likely
 				// fail, but this is desired, so we can quickly move on to a faster one - we should
 				// also decrease its reputation.
-				request_timeout: Duration::from_secs(1),
-				inbound_queue: tx,
-			},
-			Protocol::DisputeSendingV1 => RequestResponseConfig {
+				Duration::from_secs(1),
+				tx,
+			),
+			Protocol::DisputeSendingV1 => N::request_response_config(
 				name,
-				fallback_names: legacy_names,
-				max_request_size: 1_000,
+				legacy_names,
+				1_000,
 				// Responses are just confirmation, in essence not even a bit. So 100 seems
 				// plenty.
-				max_response_size: 100,
-				request_timeout: DISPUTE_REQUEST_TIMEOUT,
-				inbound_queue: tx,
-			},
-			Protocol::AttestedCandidateV2 => RequestResponseConfig {
+				100,
+				DISPUTE_REQUEST_TIMEOUT,
+				tx,
+			),
+			Protocol::AttestedCandidateV2 => N::request_response_config(
 				name,
-				fallback_names: legacy_names,
-				max_request_size: 1_000,
-				max_response_size: ATTESTED_CANDIDATE_RESPONSE_SIZE,
-				request_timeout: ATTESTED_CANDIDATE_TIMEOUT,
-				inbound_queue: tx,
-			},
+				legacy_names,
+				1_000,
+				ATTESTED_CANDIDATE_RESPONSE_SIZE,
+				ATTESTED_CANDIDATE_TIMEOUT,
+				tx,
+			),
 		}
 	}

@@ -20,7 +20,7 @@ use network::ProtocolName;
 use parity_scale_codec::{Decode, Encode, Error as DecodingError};

 use sc_network as network;
-use sc_network::PeerId;
+use sc_network_types::PeerId;

 use polkadot_primitives::AuthorityDiscoveryId;