Merge branch 'master' into ao-past-session-slashing-client

* master:
  runtime: past session slashing runtime API (#6667)
  cli: enable BEEFY by default on test networks (#7293)
  pallets: implement `Default` for `GenesisConfig` in `no_std` (#7271)
  Update quote to 1.0.27 (#7280)
  PVF: Refactor workers into separate crates, remove host dependency (#7253)

Author: ordian

Cargo.lock

@@ -24,7 +24,7 @@ tikv-jemallocator = "0.5.0"
 
 # Crates in our workspace, defined as dependencies so we can pass them feature flags.
 polkadot-cli = { path = "cli", features = [ "kusama-native", "westend-native", "rococo-native" ]  }
-polkadot-node-core-pvf-worker = { path = "node/core/pvf/worker" }
+polkadot-node-core-pvf-prepare-worker = { path = "node/core/pvf/prepare-worker" }
 polkadot-overseer = { path = "node/overseer" }
 
 [dev-dependencies]
@@ -81,7 +81,9 @@ members = [
 	"node/core/parachains-inherent",
 	"node/core/provisioner",
 	"node/core/pvf",
-	"node/core/pvf/worker",
+	"node/core/pvf/common",
+	"node/core/pvf/execute-worker",
+	"node/core/pvf/prepare-worker",
 	"node/core/pvf-checker",
 	"node/core/runtime-api",
 	"node/network/approval-distribution",
@@ -208,7 +210,7 @@ try-runtime = [ "polkadot-cli/try-runtime" ]
 fast-runtime = [ "polkadot-cli/fast-runtime" ]
 runtime-metrics = [ "polkadot-cli/runtime-metrics" ]
 pyroscope = ["polkadot-cli/pyroscope"]
-jemalloc-allocator = ["polkadot-node-core-pvf-worker/jemalloc-allocator", "polkadot-overseer/jemalloc-allocator"]
+jemalloc-allocator = ["polkadot-node-core-pvf-prepare-worker/jemalloc-allocator", "polkadot-overseer/jemalloc-allocator"]
 
 
 

Cargo.toml

@@ -23,7 +23,8 @@ pyroscope_pprofrs = { version = "0.2", optional = true }
 
 service = { package = "polkadot-service", path = "../node/service", default-features = false, optional = true }
 polkadot-client = { path = "../node/client", optional = true }
-polkadot-node-core-pvf-worker = { path = "../node/core/pvf/worker", optional = true }
+polkadot-node-core-pvf-execute-worker = { path = "../node/core/pvf/execute-worker", optional = true }
+polkadot-node-core-pvf-prepare-worker = { path = "../node/core/pvf/prepare-worker", optional = true }
 polkadot-performance-test = { path = "../node/test/performance-test", optional = true }
 
 sp-core = { git = "https://github.com/paritytech/substrate", branch = "master" }
@@ -54,7 +55,8 @@ cli = [
 	"frame-benchmarking-cli",
 	"try-runtime-cli",
 	"polkadot-client",
-	"polkadot-node-core-pvf-worker",
+	"polkadot-node-core-pvf-execute-worker",
+	"polkadot-node-core-pvf-prepare-worker",
 ]
 runtime-benchmarks = [
 	"service/runtime-benchmarks",

cli/Cargo.toml

@@ -114,9 +114,10 @@ pub struct RunCmd {
 	#[arg(long = "grandpa-pause", num_args = 2)]
 	pub grandpa_pause: Vec<u32>,
 
-	/// Enable the BEEFY gadget (only on Rococo or Wococo for now).
+	/// Disable the BEEFY gadget
+	/// (currently enabled by default on Rococo, Wococo and Versi).
 	#[arg(long)]
-	pub beefy: bool,
+	pub no_beefy: bool,
 
 	/// Add the destination address to the jaeger agent.
 	///

cli/src/cli.rs

@@ -298,12 +298,9 @@ where
 		.map_err(Error::from)?;
 	let chain_spec = &runner.config().chain_spec;
 
-	// Disallow BEEFY on production networks.
-	if cli.run.beefy &&
-		(chain_spec.is_polkadot() || chain_spec.is_kusama() || chain_spec.is_westend())
-	{
-		return Err(Error::Other("BEEFY disallowed on production networks".to_string()))
-	}
+	// By default, enable BEEFY on test networks.
+	let enable_beefy = (chain_spec.is_rococo() || chain_spec.is_wococo() || chain_spec.is_versi()) &&
+		!cli.run.no_beefy;
 
 	set_default_ss58_version(chain_spec);
 
@@ -346,7 +343,7 @@ where
 			config,
 			service::IsCollator::No,
 			grandpa_pause,
-			cli.run.beefy,
+			enable_beefy,
 			jaeger_agent,
 			None,
 			false,
@@ -495,7 +492,7 @@ pub fn run() -> Result<()> {
 
 			#[cfg(not(target_os = "android"))]
 			{
-				polkadot_node_core_pvf_worker::prepare_worker_entrypoint(
+				polkadot_node_core_pvf_prepare_worker::worker_entrypoint(
 					&cmd.socket_path,
 					Some(&cmd.node_impl_version),
 				);
@@ -517,7 +514,7 @@ pub fn run() -> Result<()> {
 
 			#[cfg(not(target_os = "android"))]
 			{
-				polkadot_node_core_pvf_worker::execute_worker_entrypoint(
+				polkadot_node_core_pvf_execute_worker::worker_entrypoint(
 					&cmd.socket_path,
 					Some(&cmd.node_impl_version),
 				);

cli/src/command.rs

@@ -4,6 +4,10 @@ version.workspace = true
 authors.workspace = true
 edition.workspace = true
 
+[[bin]]
+name = "puppet_worker"
+path = "bin/puppet_worker.rs"
+
 [dependencies]
 always-assert = "0.1"
 futures = "0.3.21"
@@ -13,12 +17,16 @@ libc = "0.2.139"
 pin-project = "1.0.9"
 rand = "0.8.5"
 slotmap = "1.0"
+tempfile = "3.3.0"
 tokio = { version = "1.24.2", features = ["fs", "process"] }
 
 parity-scale-codec = { version = "3.4.0", default-features = false, features = ["derive"] }
 
 polkadot-parachain = { path = "../../../parachain" }
 polkadot-core-primitives = { path = "../../../core-primitives" }
+polkadot-node-core-pvf-common = { path = "common" }
+polkadot-node-core-pvf-execute-worker = { path = "execute-worker" }
+polkadot-node-core-pvf-prepare-worker = { path = "prepare-worker" }
 polkadot-node-metrics = { path = "../../metrics" }
 polkadot-node-primitives = { path = "../../primitives" }
 polkadot-primitives = { path = "../../../primitives" }
@@ -34,4 +42,6 @@ substrate-build-script-utils = { git = "https://github.com/paritytech/substrate"
 [dev-dependencies]
 assert_matches = "1.4.0"
 hex-literal = "0.3.4"
-tempfile = "3.3.0"
+
+adder = { package = "test-parachain-adder", path = "../../../parachain/test-parachains/adder" }
+halt = { package = "test-parachain-halt", path = "../../../parachain/test-parachains/halt" }

node/core/pvf/Cargo.toml

@@ -14,4 +14,4 @@
 // You should have received a copy of the GNU General Public License
 // along with Polkadot.  If not, see <http://www.gnu.org/licenses/>.
 
-polkadot_node_core_pvf_worker::decl_puppet_worker_main!();
+polkadot_node_core_pvf::decl_puppet_worker_main!();

node/core/pvf/worker/bin/puppet_worker.rs renamed to node/core/pvf/bin/puppet_worker.rs

@@ -0,0 +1,26 @@
+[package]
+name = "polkadot-node-core-pvf-common"
+version.workspace = true
+authors.workspace = true
+edition.workspace = true
+
+[dependencies]
+cpu-time = "1.0.0"
+futures = "0.3.21"
+gum = { package = "tracing-gum", path = "../../../gum" }
+libc = "0.2.139"
+tokio = { version = "1.24.2", features = ["fs", "process", "io-util"] }
+
+parity-scale-codec = { version = "3.4.0", default-features = false, features = ["derive"] }
+
+polkadot-parachain = { path = "../../../../parachain" }
+polkadot-primitives = { path = "../../../../primitives" }
+
+sc-executor-common = { git = "https://github.com/paritytech/substrate", branch = "master" }
+sc-executor-wasmtime = { git = "https://github.com/paritytech/substrate", branch = "master" }
+
+sp-core = { git = "https://github.com/paritytech/substrate", branch = "master" }
+sp-tracing = { git = "https://github.com/paritytech/substrate", branch = "master" }
+
+[build-dependencies]
+substrate-build-script-utils = { git = "https://github.com/paritytech/substrate", branch = "master" }

node/core/pvf/common/Cargo.toml

@@ -0,0 +1,106 @@
+// 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/>.
+
+use crate::prepare::PrepareStats;
+use parity_scale_codec::{Decode, Encode};
+use std::fmt;
+
+/// Result of PVF preparation performed by the validation host. Contains stats about the preparation if
+/// successful
+pub type PrepareResult = Result<PrepareStats, PrepareError>;
+
+/// An error that occurred during the prepare part of the PVF pipeline.
+#[derive(Debug, Clone, Encode, Decode)]
+pub enum PrepareError {
+	/// During the prevalidation stage of preparation an issue was found with the PVF.
+	Prevalidation(String),
+	/// Compilation failed for the given PVF.
+	Preparation(String),
+	/// An unexpected panic has occurred in the preparation worker.
+	Panic(String),
+	/// Failed to prepare the PVF due to the time limit.
+	TimedOut,
+	/// An IO error occurred. This state is reported by either the validation host or by the worker.
+	IoErr(String),
+	/// The temporary file for the artifact could not be created at the given cache path. This state is reported by the
+	/// validation host (not by the worker).
+	CreateTmpFileErr(String),
+	/// The response from the worker is received, but the file cannot be renamed (moved) to the final destination
+	/// location. This state is reported by the validation host (not by the worker).
+	RenameTmpFileErr(String),
+}
+
+impl PrepareError {
+	/// Returns whether this is a deterministic error, i.e. one that should trigger reliably. Those
+	/// errors depend on the PVF itself and the sc-executor/wasmtime logic.
+	///
+	/// Non-deterministic errors can happen spuriously. Typically, they occur due to resource
+	/// starvation, e.g. under heavy load or memory pressure. Those errors are typically transient
+	/// but may persist e.g. if the node is run by overwhelmingly underpowered machine.
+	pub fn is_deterministic(&self) -> bool {
+		use PrepareError::*;
+		match self {
+			Prevalidation(_) | Preparation(_) | Panic(_) => true,
+			TimedOut | IoErr(_) | CreateTmpFileErr(_) | RenameTmpFileErr(_) => false,
+		}
+	}
+}
+
+impl fmt::Display for PrepareError {
+	fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+		use PrepareError::*;
+		match self {
+			Prevalidation(err) => write!(f, "prevalidation: {}", err),
+			Preparation(err) => write!(f, "preparation: {}", err),
+			Panic(err) => write!(f, "panic: {}", err),
+			TimedOut => write!(f, "prepare: timeout"),
+			IoErr(err) => write!(f, "prepare: io error while receiving response: {}", err),
+			CreateTmpFileErr(err) => write!(f, "prepare: error creating tmp file: {}", err),
+			RenameTmpFileErr(err) => write!(f, "prepare: error renaming tmp file: {}", err),
+		}
+	}
+}
+
+/// Some internal error occurred.
+///
+/// Should only ever be used for validation errors independent of the candidate and PVF, or for errors we ruled out
+/// during pre-checking (so preparation errors are fine).
+#[derive(Debug, Clone, Encode, Decode)]
+pub enum InternalValidationError {
+	/// Some communication error occurred with the host.
+	HostCommunication(String),
+	/// Could not find or open compiled artifact file.
+	CouldNotOpenFile(String),
+	/// An error occurred in the CPU time monitor thread. Should be totally unrelated to validation.
+	CpuTimeMonitorThread(String),
+	/// Some non-deterministic preparation error occurred.
+	NonDeterministicPrepareError(PrepareError),
+}
+
+impl fmt::Display for InternalValidationError {
+	fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+		use InternalValidationError::*;
+		match self {
+			HostCommunication(err) =>
+				write!(f, "validation: some communication error occurred with the host: {}", err),
+			CouldNotOpenFile(err) =>
+				write!(f, "validation: could not find or open compiled artifact file: {}", err),
+			CpuTimeMonitorThread(err) =>
+				write!(f, "validation: an error occurred in the CPU time monitor thread: {}", err),
+			NonDeterministicPrepareError(err) => write!(f, "validation: prepare: {}", err),
+		}
+	}
+}

node/core/pvf/worker/build.rs renamed to node/core/pvf/common/build.rs

@@ -0,0 +1,60 @@
+// 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/>.
+
+use crate::error::InternalValidationError;
+use parity_scale_codec::{Decode, Encode};
+use polkadot_parachain::primitives::ValidationResult;
+use polkadot_primitives::ExecutorParams;
+use std::time::Duration;
+
+/// The payload of the one-time handshake that is done when a worker process is created. Carries
+/// data from the host to the worker.
+#[derive(Encode, Decode)]
+pub struct Handshake {
+	/// The executor parameters.
+	pub executor_params: ExecutorParams,
+}
+
+/// The response from an execution job on the worker.
+#[derive(Encode, Decode)]
+pub enum Response {
+	/// The job completed successfully.
+	Ok {
+		/// The result of parachain validation.
+		result_descriptor: ValidationResult,
+		/// The amount of CPU time taken by the job.
+		duration: Duration,
+	},
+	/// The candidate is invalid.
+	InvalidCandidate(String),
+	/// The job timed out.
+	TimedOut,
+	/// An unexpected panic has occurred in the execution worker.
+	Panic(String),
+	/// Some internal error occurred.
+	InternalError(InternalValidationError),
+}
+
+impl Response {
+	/// Creates an invalid response from a context `ctx` and a message `msg` (which can be empty).
+	pub fn format_invalid(ctx: &'static str, msg: &str) -> Self {
+		if msg.is_empty() {
+			Self::InvalidCandidate(ctx.to_string())
+		} else {
+			Self::InvalidCandidate(format!("{}: {}", ctx, msg))
+		}
+	}
+}