Changeset - dd4e6a5314f7
[Not reviewed]
0 6 0
MH - 4 years ago 2021-11-23 15:16:34
contact@maxhenger.nl
WIP on more failure fixing
6 files changed with 164 insertions and 35 deletions:
0 comments (0 inline, 0 general)
src/collections/sets.rs
Show inline comments
 
@@ -61,53 +61,58 @@ impl<T: Eq> DequeSet<T> {
 
pub struct VecSet<T: Eq> {
 
    inner: Vec<T>,
 
}
 

	
 
impl<T: Eq> VecSet<T> {
 
    pub fn new() -> Self {
 
        Self{ inner: Vec::new() }
 
    }
 

	
 
    #[inline]
 
    pub fn pop(&mut self) -> Option<T> {
 
        self.inner.pop()
 
    }
 

	
 
    /// Pushes a new element into the set. Returns `false` if it was already
 
    /// present and `true` if it is newly added.
 
    #[inline]
 
    pub fn push(&mut self, to_push: T) -> bool {
 
        for element in self.inner.iter() {
 
            if *element == to_push {
 
                return false;
 
            }
 
        }
 

	
 
        self.inner.push(to_push);
 
        return true
 
    }
 

	
 
    #[inline]
 
    pub fn clear(&mut self) {
 
        self.inner.clear();
 
    }
 

	
 
    #[inline]
 
    pub fn iter(&self) -> impl Iterator<Item=&T> {
 
        return self.inner.iter();
 
    }
 

	
 
    #[inline]
 
    pub fn contains(&self, item: &T) -> bool {
 
        return self.inner.contains(item);
 
    }
 

	
 
    #[inline]
 
    pub fn is_empty(&self) -> bool {
 
        self.inner.is_empty()
 
    }
 

	
 
    #[inline]
 
    pub fn len(&self) -> usize {
 
        return self.inner.len();
 
    }
 

	
 
    #[inline]
 
    pub fn into_vec(self) -> Vec<T> {
 
        return self.inner;
 
    }
 
}
 
\ No newline at end of file
src/runtime2/connector.rs
Show inline comments
 
@@ -358,102 +358,104 @@ impl ConnectorPDL {
 
                    return ConnectorScheduling::Immediate;
 
                }
 
            }
 
            EvalContinuation::SyncBlockEnd => {
 
                let consistency = self.consensus.notify_of_finished_branch(branch_id);
 
                if consistency == Consistency::Valid {
 
                    branch.sync_state = SpeculativeState::ReachedSyncEnd;
 
                    self.tree.push_into_queue(QueueKind::FinishedSync, branch_id);
 
                } else {
 
                    branch.sync_state = SpeculativeState::Inconsistent;
 
                }
 
            },
 
            EvalContinuation::NewFork => {
 
                // Like the `NewChannel` result. This means we're setting up
 
                // a branch and putting a marker inside the RunContext for the
 
                // next time we run the PDL code
 
                let left_id = branch_id;
 
                let right_id = self.tree.fork_branch(left_id);
 
                self.consensus.notify_of_new_branch(left_id, right_id);
 
                self.tree.push_into_queue(QueueKind::Runnable, left_id);
 
                self.tree.push_into_queue(QueueKind::Runnable, right_id);
 

	
 
                let left_branch = &mut self.tree[left_id];
 
                left_branch.prepared = PreparedStatement::ForkedExecution(true);
 
                let right_branch = &mut self.tree[right_id];
 
                right_branch.prepared = PreparedStatement::ForkedExecution(false);
 
            }
 
            EvalContinuation::Put(port_id, content) => {
 
                // Branch is attempting to send data
 
                let port_id = PortIdLocal::new(port_id.0.u32_suffix);
 
                let (sync_header, data_header) = self.consensus.handle_message_to_send(branch_id, port_id, &content, comp_ctx);
 
                let message = DataMessage{ sync_header, data_header, content };
 
                match comp_ctx.submit_message(Message::Data(message)) {
 
                    Ok(_) => {
 
                        // Message is underway
 
                        branch.prepared = PreparedStatement::PerformedPut;
 
                        self.tree.push_into_queue(QueueKind::Runnable, branch_id);
 
                        return ConnectorScheduling::Immediate;
 
                    },
 
                    Err(_) => {
 
                        // We don't own the port
 
                        let pd = &sched_ctx.runtime.protocol_description;
 
                        let eval_error = branch.code_state.new_error_at_expr(
 
                            &pd.modules, &pd.heap,
 
                            String::from("attempted to 'put' on port that is no longer owned")
 
                        );
 
                        self.eval_error = Some(eval_error);
 
                        self.mode = Mode::SyncError;
 

	
 
                        println!("DEBUGERINO: Notify of fatal branch");
 
                        if let Some(conclusion) = self.consensus.notify_of_fatal_branch(branch_id, comp_ctx) {
 
                            println!("DEBUGERINO: Actually got {:?}", conclusion);
 
                            return self.enter_non_sync_mode(conclusion, comp_ctx);
 
                        }
 
                    }
 
                }
 

	
 
                branch.prepared = PreparedStatement::PerformedPut;
 
                self.tree.push_into_queue(QueueKind::Runnable, branch_id);
 
                return ConnectorScheduling::Immediate;
 
            },
 
            _ => unreachable!("unexpected run result {:?} in sync mode", run_result),
 
        }
 

	
 
        // If here then the run result did not require a particular action. We
 
        // return whether we have more active branches to run or not.
 
        if self.tree.queue_is_empty(QueueKind::Runnable) {
 
            return ConnectorScheduling::NotNow;
 
        } else {
 
            return ConnectorScheduling::Later;
 
        }
 
    }
 

	
 
    pub fn run_in_deterministic_mode(&mut self, sched_ctx: SchedulerCtx, comp_ctx: &mut ComponentCtx) -> ConnectorScheduling {
 
        debug_assert!(!self.tree.is_in_sync() && !self.consensus.is_in_sync());
 

	
 
        let branch = self.tree.base_branch_mut();
 
        debug_assert!(branch.sync_state == SpeculativeState::RunningNonSync);
 

	
 
        let mut run_context = ConnectorRunContext{
 
            branch_id: branch.id,
 
            consensus: &self.consensus,
 
            prepared: branch.prepared.take(),
 
        };
 
        let run_result = Self::run_prompt(&mut branch.code_state, &sched_ctx.runtime.protocol_description, &mut run_context);
 
        if let Err(eval_error) = run_result {
 
            comp_ctx.push_error(eval_error);
 
            return ConnectorScheduling::Exit
 
        }
 
        let run_result = run_result.unwrap();
 

	
 
        match run_result {
 
            EvalContinuation::ComponentTerminated => {
 
                branch.sync_state = SpeculativeState::Finished;
 
                return ConnectorScheduling::Exit;
 
            },
 
            EvalContinuation::SyncBlockStart => {
 
                comp_ctx.notify_sync_start();
 
                let sync_branch_id = self.tree.start_sync();
 
                debug_assert!(self.last_finished_handled.is_none());
 
                self.consensus.start_sync(comp_ctx);
 
                self.consensus.notify_of_new_branch(BranchId::new_invalid(), sync_branch_id);
 
                self.tree.push_into_queue(QueueKind::Runnable, sync_branch_id);
 
                self.mode = Mode::Sync;
 

	
 
                return ConnectorScheduling::Immediate;
 
            },
 
            EvalContinuation::NewComponent(definition_id, monomorph_idx, arguments) => {
src/runtime2/consensus.rs
Show inline comments
 
use crate::collections::VecSet;
 

	
 
use crate::protocol::eval::ValueGroup;
 
use crate::runtime2::inbox::SyncCompContent::Presence;
 
use crate::runtime2::port::PortState;
 

	
 
use super::ConnectorId;
 
use super::branch::BranchId;
 
use super::port::{ChannelId, PortIdLocal};
 
use super::inbox::{
 
    Message, DataHeader, SyncHeader, ChannelAnnotation, BranchMarker,
 
    DataMessage,
 
    SyncCompMessage, SyncCompContent,
 
    SyncPortMessage, SyncPortContent,
 
    SyncControlMessage, SyncControlContent
 
};
 
use super::scheduler::{ComponentCtx, ComponentPortChange};
 

	
 
struct BranchAnnotation {
 
    channel_mapping: Vec<ChannelAnnotation>,
 
    cur_marker: BranchMarker,
 
}
 

	
 
#[derive(Debug)]
 
pub(crate) struct LocalSolution {
 
    component: ConnectorId,
 
    final_branch_id: BranchId,
 
    port_mapping: Vec<(ChannelId, BranchMarker)>,
 
}
 

	
 
#[derive(Debug, Clone)]
 
pub(crate) struct GlobalSolution {
 
    component_branches: Vec<(ConnectorId, BranchId)>,
 
    channel_mapping: Vec<(ChannelId, BranchMarker)>, // TODO: This can go, is debugging info
 
}
 

	
 
#[derive(Debug, PartialEq, Eq)]
 
pub enum RoundConclusion {
 
    Failure,
 
    Success(BranchId),
 
}
 

	
 
// -----------------------------------------------------------------------------
 
// Consensus
 
// -----------------------------------------------------------------------------
 

	
 
struct Peer {
 
    id: ConnectorId,
 
    encountered_this_round: bool,
 
    expected_sync_round: u32,
 
}
 

	
 
/// The consensus algorithm. Currently only implemented to find the component
 
@@ -122,103 +124,118 @@ impl Consensus {
 
    pub fn start_sync(&mut self, ctx: &ComponentCtx) {
 
        debug_assert!(!self.highest_connector_id.is_valid());
 
        debug_assert!(self.branch_annotations.is_empty());
 
        debug_assert!(self.solution_combiner.local.is_empty());
 

	
 
        // We'll use the first "branch" (the non-sync one) to store our ports,
 
        // this allows cloning if we created a new branch.
 
        self.branch_annotations.push(BranchAnnotation{
 
            channel_mapping: ctx.get_ports().iter()
 
                .map(|v| ChannelAnnotation {
 
                    channel_id: v.channel_id,
 
                    registered_id: None,
 
                    expected_firing: None,
 
                })
 
                .collect(),
 
            cur_marker: BranchMarker::new_invalid(),
 
        });
 
        self.branch_markers.push(BranchId::new_invalid());
 

	
 
        self.highest_connector_id = ctx.id;
 

	
 
    }
 

	
 
    /// Notifies the consensus algorithm that a new branch has appeared. Must be
 
    /// called for each forked branch in the execution tree.
 
    pub fn notify_of_new_branch(&mut self, parent_branch_id: BranchId, new_branch_id: BranchId) {
 
        // If called correctly. Then each time we are notified the new branch's
 
        // index is the length in `branch_annotations`.
 
        debug_assert!(self.branch_annotations.len() == new_branch_id.index as usize);
 
        let parent_branch_annotations = &self.branch_annotations[parent_branch_id.index as usize];
 
        let new_marker = BranchMarker::new(self.branch_markers.len() as u32);
 
        let new_branch_annotations = BranchAnnotation{
 
            channel_mapping: parent_branch_annotations.channel_mapping.clone(),
 
            cur_marker: new_marker,
 
        };
 
        self.branch_annotations.push(new_branch_annotations);
 
        self.branch_markers.push(new_branch_id);
 
    }
 

	
 
    /// Notifies the consensus algorithm that a particular branch has
 
    /// encountered an unrecoverable error.
 
    pub fn notify_of_fatal_branch(&mut self, failed_branch_id: BranchId, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        debug_assert!(self.is_in_sync());
 

	
 
        // Check for trivial case, where branch has not yet communicated within
 
        // the consensus algorithm
 
        let branch = &self.branch_annotations[failed_branch_id.index as usize];
 
        if branch.channel_mapping.iter().all(|v| v.registered_id.is_none()) {
 
            println!("DEBUG: Failure everything silent");
 
            return Some(RoundConclusion::Failure);
 
        }
 

	
 
        // We need to go through the hassle of notifying all participants in the
 
        // sync round that we've encountered an error.
 
        // --- notify leader
 
        let mut channel_presence = Vec::with_capacity(branch.channel_mapping.len());
 
        for mapping in &branch.channel_mapping {
 
            let port = ctx.get_port_by_channel_id(mapping.channel_id).unwrap();
 
            channel_presence.push(LocalChannelPresence{
 
                channel_id: mapping.channel_id,
 
                is_closed: port.state == PortState::Closed,
 
            });
 
        }
 
        let _never_conclusion = self.send_to_leader_or_handle_as_leader(SyncCompContent::Presence(ComponentPresence{
 
            component_id: ctx.id,
 
            channels: channel_presence,
 
        }), ctx);
 
        debug_assert!(_never_conclusion.is_none());
 
        let maybe_conclusion = self.send_to_leader_or_handle_as_leader(SyncCompContent::LocalFailure, ctx);
 
        println!("DEBUG: Maybe conclusion is {:?}", maybe_conclusion);
 

	
 
        // --- initiate discovery wave (to let leader know about all components)
 
        self.handled_wave = true;
 
        for mapping in &self.branch_annotations[0].channel_mapping {
 
            let channel_id = mapping.channel_id;
 
            let port_info = ctx.get_port_by_channel_id(channel_id).unwrap();
 
            let message = SyncPortMessage{
 
                sync_header: self.create_sync_header(ctx),
 
                source_port: port_info.self_id,
 
                target_port: port_info.peer_id,
 
                content: SyncPortContent::NotificationWave,
 
            };
 

	
 
            // Note: submitting the message might fail. But we're attempting to
 
            // handle the error anyway.
 
            // TODO: Think about this a second time: how do we make sure the
 
            //  entire network will fail if we reach this condition
 
            let _unused = ctx.submit_message(Message::SyncPort(message));
 
        }
 

	
 
        return maybe_conclusion;
 
    }
 

	
 
    /// Notifies the consensus algorithm that a branch has reached the end of
 
    /// the sync block. A final check for consistency will be performed that the
 
    /// caller has to handle. Note that
 
    pub fn notify_of_finished_branch(&self, branch_id: BranchId) -> Consistency {
 
        debug_assert!(self.is_in_sync());
 
        let branch = &self.branch_annotations[branch_id.index as usize];
 
        for mapping in &branch.channel_mapping {
 
            match mapping.expected_firing {
 
                Some(expected) => {
 
                    if expected != mapping.registered_id.is_some() {
 
                        // Inconsistent speculative state and actual state
 
                        debug_assert!(mapping.registered_id.is_none()); // because if we did fire on a silent port, we should've caught that earlier
 
                        return Consistency::Inconsistent;
 
                    }
 
                },
 
                None => {},
 
            }
 
        }
 

	
 
        return Consistency::Valid;
 
    }
 

	
 
    /// Notifies the consensus algorithm that a particular branch has assumed
 
    /// a speculative value for its port mapping.
 
    pub fn notify_of_speculative_mapping(&mut self, branch_id: BranchId, port_id: PortIdLocal, does_fire: bool, ctx: &ComponentCtx) -> Consistency {
 
@@ -373,177 +390,198 @@ impl Consensus {
 
            new_mapping: branch.cur_marker,
 
        };
 

	
 
        // Update port mapping
 
        for mapping in &mut branch.channel_mapping {
 
            if mapping.channel_id == port_info.channel_id {
 
                mapping.expected_firing = Some(true);
 
                mapping.registered_id = Some(branch.cur_marker);
 
            }
 
        }
 

	
 
        // Update branch marker
 
        let new_marker = BranchMarker::new(self.branch_markers.len() as u32);
 
        branch.cur_marker = new_marker;
 
        self.branch_markers.push(branch_id);
 

	
 
        self.encountered_ports.push(source_port_id);
 

	
 
        return (self.create_sync_header(ctx), data_header);
 
    }
 

	
 
    /// Handles a new data message by handling the sync header. The caller is
 
    /// responsible for checking for branches that might be able to receive
 
    /// the message.
 
    pub fn handle_new_data_message(&mut self, message: &DataMessage, ctx: &mut ComponentCtx) -> bool {
 
        let handled = self.handle_received_sync_header(&message.sync_header, ctx);
 
        if handled {
 
            self.encountered_ports.push(message.data_header.target_port);
 
        }
 
        return handled;
 
    }
 

	
 
    /// Handles a new sync message by handling the sync header and the contents
 
    /// of the message. Returns `Some` with the branch ID of the global solution
 
    /// if the sync solution has been found.
 
    pub fn handle_new_sync_comp_message(&mut self, message: SyncCompMessage, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        if !self.handle_received_sync_header(&message.sync_header, ctx) {
 
            return None;
 
        }
 

	
 
        // And handle the contents
 
        debug_assert_eq!(message.target_component_id, ctx.id);
 

	
 
        match &message.content {
 
            SyncCompContent::LocalFailure |
 
            SyncCompContent::LocalSolution(_) |
 
            SyncCompContent::PartialSolution(_) |
 
            SyncCompContent::AckFailure |
 
            SyncCompContent::Presence(_, _) => {
 
            SyncCompContent::Presence(_) => {
 
                // Needs to be handled by the leader
 
                return self.send_to_leader_or_handle_as_leader(message.content, ctx);
 
            },
 
            SyncCompContent::GlobalSolution(solution) => {
 
                // Found a global solution
 
                debug_assert_ne!(self.highest_connector_id, ctx.id); // not the leader
 
                let (_, branch_id) = solution.component_branches.iter()
 
                    .find(|(component_id, _)| *component_id == ctx.id)
 
                    .unwrap();
 
                return Some(RoundConclusion::Success(*branch_id));
 
            },
 
            SyncCompContent::GlobalFailure => {
 
                // Global failure of round, send Ack to leader
 
                println!("DEBUGERINO: Got GlobalFailure, sending Ack in response");
 
                debug_assert_ne!(self.highest_connector_id, ctx.id); // not the leader
 
                let _result = self.send_to_leader_or_handle_as_leader(SyncCompContent::AckFailure, ctx);
 
                debug_assert!(_result.is_none());
 
                return Some(RoundConclusion::Failure);
 
            },
 
            SyncCompContent::Notification => {
 
                // We were just interested in the sync header we handled above
 
                return None;
 
            }
 
        }
 
    }
 

	
 
    pub fn handle_new_sync_port_message(&mut self, message: SyncPortMessage, ctx: &mut ComponentCtx) {
 
    pub fn handle_new_sync_port_message(&mut self, message: SyncPortMessage, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        if !self.handle_received_sync_header(&message.sync_header, ctx) {
 
            return;
 
            return None;
 
        }
 

	
 
        debug_assert!(self.is_in_sync());
 
        debug_assert!(ctx.get_port_by_id(message.target_port).is_some());
 
        match message.content {
 
            SyncPortContent::SilentPortNotification => {
 
                // The point here is to let us become part of the sync round and
 
                // take note of the leader in case all of our ports are silent.
 
                self.encountered_ports.push(message.target_port);
 
                return None
 
            }
 
            SyncPortContent::NotificationWave => {
 
                // Wave to discover everyone in the network, handling sync
 
                // header takes care of leader discovery, here we need to make
 
                // sure we propagate the wave
 
                if self.handled_wave {
 
                    return;
 
                    return None;
 
                }
 

	
 
                self.handled_wave = true;
 

	
 
                // Propagate wave to all peers except the one that has sent us
 
                // the wave.
 
                for mapping in &self.branch_annotations[0].channel_mapping {
 
                    let channel_id = mapping.channel_id;
 
                    let port_desc = ctx.get_port_by_channel_id(channel_id).unwrap();
 
                    if port_desc.self_id == message.target_port {
 
                        // Wave came from this port, no need to send one back
 
                        continue;
 
                    }
 

	
 
                    let message = SyncPortMessage{
 
                        sync_header: self.create_sync_header(ctx),
 
                        source_port: port_desc.self_id,
 
                        target_port: port_desc.peer_id,
 
                        content: SyncPortContent::NotificationWave,
 
                    };
 
                    // As with the other SyncPort where we throw away the
 
                    // result: we're dealing with an error here anyway
 
                    let _unused = ctx.submit_message(Message::SyncPort(message));
 
                }
 

	
 
                // And let the leader know about our port state
 
                let annotations = &self.branch_annotations[0];
 
                let mut channels = Vec::with_capacity(annotations.channel_mapping.len());
 
                for mapping in &annotations.channel_mapping {
 
                    let port_info = ctx.get_port_by_channel_id(mapping.channel_id).unwrap();
 
                    channels.push(LocalChannelPresence{
 
                        channel_id: mapping.channel_id,
 
                        is_closed: port_info.state == PortState::Closed,
 
                    });
 
                }
 

	
 
                let maybe_conlusion = self.send_to_leader_or_handle_as_leader(SyncCompContent::Presence(ComponentPresence{
 
                    component_id: ctx.id,
 
                    channels,
 
                }), ctx);
 
                return maybe_conlusion;
 
            }
 
        }
 
    }
 

	
 
    pub fn handle_new_sync_control_message(&mut self, message: SyncControlMessage, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        if message.in_response_to_sync_round < self.sync_round {
 
            // Old message
 
            return None
 
        }
 

	
 
        match message.content {
 
            SyncControlContent::ChannelIsClosed(_) => {
 
                // TODO: This is wrong! This might happen in a normal sync. And
 
                // we don't want to fail immediately!
 
                return Some(RoundConclusion::Failure);
 
            }
 
        }
 
    }
 

	
 
    pub fn notify_of_received_message(&mut self, branch_id: BranchId, message: &DataMessage, ctx: &ComponentCtx) {
 
        debug_assert!(self.branch_can_receive(branch_id, message));
 

	
 
        let target_port = ctx.get_port_by_id(message.data_header.target_port).unwrap();
 
        let branch = &mut self.branch_annotations[branch_id.index as usize];
 
        for mapping in &mut branch.channel_mapping {
 
            if mapping.channel_id == target_port.channel_id {
 
                // Found the port in which the message should be inserted
 
                mapping.registered_id = Some(message.data_header.new_mapping);
 

	
 
                // Check for sent ports
 
                debug_assert!(self.workspace_ports.is_empty());
 
                find_ports_in_value_group(&message.content, &mut self.workspace_ports);
 
                if !self.workspace_ports.is_empty() {
 
                    todo!("handle received ports");
 
                    self.workspace_ports.clear();
 
                }
 

	
 
                return;
 
            }
 
        }
 

	
 
        // If here, then the branch didn't actually own the port? Means the
 
        // caller made a mistake
 
        unreachable!("incorrect notify_of_received_message");
 
    }
 

	
 
    /// Matches the mapping between the branch and the data message. If they
 
    /// match then the branch can receive the message.
 
    pub fn branch_can_receive(&self, branch_id: BranchId, message: &DataMessage) -> bool {
 
        if let Some(peer) = self.peers.iter().find(|v| v.id == message.sync_header.sending_component_id) {
 
            if message.sync_header.sync_round < peer.expected_sync_round {
 
                return false;
 
            }
 
        }
 

	
 
        let annotation = &self.branch_annotations[branch_id.index as usize];
 
        for expected in &message.data_header.expected_mapping {
 
            // If we own the port, then we have an entry in the
 
            // annotation, check if the current mapping matches
 
            for current in &annotation.channel_mapping {
 
                if expected.channel_id == current.channel_id {
 
                    if expected.registered_id != current.registered_id {
 
@@ -611,249 +649,288 @@ impl Consensus {
 
                entry.encountered_this_round = true;
 
                // TODO: Proper handling of potential overflow
 
                if sync_header.sync_round >= entry.expected_sync_round {
 
                    entry.expected_sync_round = sync_header.sync_round;
 
                    return true;
 
                } else {
 
                    return false;
 
                }
 
            },
 
            None => {
 
                self.peers.push(Peer{
 
                    id: sync_header.sending_component_id,
 
                    encountered_this_round: true,
 
                    expected_sync_round: sync_header.sync_round,
 
                });
 
                return true;
 
            }
 
        }
 
    }
 

	
 
    /// Sends a message towards the leader, if already the leader then the
 
    /// message will be handled immediately.
 
    fn send_to_leader_or_handle_as_leader(&mut self, content: SyncCompContent, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        if self.highest_connector_id == ctx.id {
 
            // We are the leader
 
            match content {
 
                SyncCompContent::LocalFailure => {
 
                    if self.solution_combiner.mark_failure_and_check_for_global_failure() {
 
                        return self.handle_global_failure_as_leader(ctx);
 
                    }
 
                },
 
                SyncCompContent::LocalSolution(local_solution) => {
 
                    if let Some(global_solution) = self.solution_combiner.add_solution_and_check_for_global_solution(local_solution) {
 
                        return self.handle_global_solution_as_leader(global_solution, ctx);
 
                    }
 
                },
 
                SyncCompContent::PartialSolution(partial_solution) => {
 
                    if let Some(conclusion) = self.solution_combiner.combine(partial_solution) {
 
                        match conclusion {
 
                            LeaderConclusion::Solution(global_solution) => {
 
                                return self.handle_global_solution_as_leader(global_solution, ctx);
 
                            },
 
                            LeaderConclusion::Failure => {
 
                                return self.handle_global_failure_as_leader(ctx);
 
                            }
 
                        }
 
                    }
 
                },
 
                SyncCompContent::Presence(component_id, presence) => {
 
                    if self.solution_combiner.add_presence_and_check_for_global_failure(component_id, &presence) {
 
                SyncCompContent::Presence(component_presence) => {
 
                    if self.solution_combiner.add_presence_and_check_for_global_failure(component_presence.component_id, &component_presence.channels) {
 
                        return self.handle_global_failure_as_leader(ctx);
 
                    }
 
                },
 
                SyncCompContent::AckFailure => {
 
                    debug_assert_eq!(Some(RoundConclusion::Failure), self.conclusion);
 
                    debug_assert!(self.ack_remaining > 0);
 
                    self.ack_remaining -= 1;
 
                    if self.ack_remaining == 0 {
 
                        return Some(RoundConclusion::Failure);
 
                    }
 
                }
 
                SyncCompContent::Notification | SyncCompContent::GlobalSolution(_) |
 
                SyncCompContent::GlobalFailure => {
 
                    unreachable!("unexpected message content for leader");
 
                },
 
            }
 
        } else {
 
            // Someone else is the leader
 
            let message = SyncCompMessage {
 
                sync_header: self.create_sync_header(ctx),
 
                target_component_id: self.highest_connector_id,
 
                content,
 
            };
 
            ctx.submit_message(Message::SyncComp(message)).unwrap(); // unwrap: sending to component instead of through channel
 
        }
 

	
 
        return None;
 
    }
 

	
 
    fn handle_global_solution_as_leader(&mut self, global_solution: GlobalSolution, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        if self.conclusion.is_some() {
 
            return None;
 
        }
 

	
 
        // Handle the global solution
 
        let mut my_final_branch_id = BranchId::new_invalid();
 
        for (connector_id, branch_id) in global_solution.component_branches.iter().copied() {
 
            if connector_id == ctx.id {
 
                // This is our solution branch
 
                my_final_branch_id = branch_id;
 
                continue;
 
            }
 

	
 
            let message = SyncCompMessage {
 
                sync_header: self.create_sync_header(ctx),
 
                target_component_id: connector_id,
 
                content: SyncCompContent::GlobalSolution(global_solution.clone()),
 
            };
 
            ctx.submit_message(Message::SyncComp(message)).unwrap(); // unwrap: sending to component instead of through channel
 
        }
 

	
 
        debug_assert!(my_final_branch_id.is_valid());
 
        self.conclusion = Some(RoundConclusion::Success(my_final_branch_id));
 
        return Some(RoundConclusion::Success(my_final_branch_id));
 
    }
 

	
 
    fn handle_global_failure_as_leader(&mut self, ctx: &mut ComponentCtx) -> Option<RoundConclusion> {
 
        debug_assert!(self.solution_combiner.failure_reported && self.solution_combiner.check_for_global_failure());
 
        if self.conclusion.is_some() {
 
            // Already sent out a failure
 
            return None;
 
        }
 

	
 
        // TODO: Performance
 
        let mut encountered = VecSet::new();
 
        for presence in &self.solution_combiner.presence {
 
            if presence.added_by != ctx.id {
 
            if presence.owner_a != ctx.id {
 
                // Did not add it ourselves
 
                if encountered.push(presence.added_by) {
 
                if encountered.push(presence.owner_a) {
 
                    // Not yet sent a message
 
                    let message = SyncCompMessage{
 
                        sync_header: self.create_sync_header(ctx),
 
                        target_component_id: presence.added_by,
 
                        target_component_id: presence.owner_a,
 
                        content: SyncCompContent::GlobalFailure,
 
                    };
 
                    ctx.submit_message(Message::SyncComp(message)).unwrap(); // unwrap: sending to component instead of through channel
 
                }
 
            } else if let Some(owner_b) = presence.owner_b {
 
                if owner_b != ctx.id {
 
                    if encountered.push(owner_b) {
 
                        let message = SyncCompMessage{
 
                            sync_header: self.create_sync_header(ctx),
 
                            target_component_id: owner_b,
 
                            content: SyncCompContent::GlobalFailure,
 
                        };
 
                        ctx.submit_message(Message::SyncComp(message)).unwrap();
 
                    }
 
                }
 
            }
 
        }
 

	
 
        println!("DEBUERINO: Leader entering error state, we need to wait on {:?}", encountered.iter().map(|v| v.0).collect::<Vec<_>>());
 
        self.conclusion = Some(RoundConclusion::Failure);
 
        if encountered.is_empty() {
 
            // We don't have to wait on Acks
 
            return Some(RoundConclusion::Failure);
 
        } else {
 
            self.ack_remaining = encountered.len() as u32;
 
            return None;
 
        }
 
    }
 

	
 
    #[inline]
 
    fn create_sync_header(&self, ctx: &ComponentCtx) -> SyncHeader {
 
        return SyncHeader{
 
            sending_component_id: ctx.id,
 
            highest_component_id: self.highest_connector_id,
 
            sync_round: self.sync_round,
 
        }
 
    }
 

	
 
    fn forward_local_data_to_new_leader(&mut self, ctx: &mut ComponentCtx) {
 
        debug_assert_ne!(self.highest_connector_id, ctx.id);
 

	
 
        if let Some(partial_solution) = self.solution_combiner.drain() {
 
            let message = SyncCompMessage {
 
                sync_header: self.create_sync_header(ctx),
 
                target_component_id: self.highest_connector_id,
 
                content: SyncCompContent::PartialSolution(partial_solution),
 
            };
 
            ctx.submit_message(Message::SyncComp(message)).unwrap(); // unwrap: sending to component instead of through channel
 
        }
 
    }
 
}
 

	
 
// -----------------------------------------------------------------------------
 
// Solution storage and algorithms
 
// -----------------------------------------------------------------------------
 

	
 
// TODO: Remove all debug derives
 

	
 
#[derive(Debug, Clone)]
 
struct MatchedLocalSolution {
 
    final_branch_id: BranchId,
 
    channel_mapping: Vec<(ChannelId, BranchMarker)>,
 
    matches: Vec<ComponentMatches>,
 
}
 

	
 
#[derive(Debug, Clone)]
 
struct ComponentMatches {
 
    target_id: ConnectorId,
 
    target_index: usize,
 
    match_indices: Vec<usize>, // of local solution in connector
 
}
 

	
 
#[derive(Debug, Clone)]
 
struct ComponentPeer {
 
    target_id: ConnectorId,
 
    target_index: usize, // in array of global solution components
 
    involved_channels: Vec<ChannelId>,
 
}
 

	
 
#[derive(Debug, Clone)]
 
struct ComponentLocalSolutions {
 
    component: ConnectorId,
 
    peers: Vec<ComponentPeer>,
 
    solutions: Vec<MatchedLocalSolution>,
 
    all_peers_present: bool,
 
}
 

	
 
#[derive(Debug, Clone)]
 
pub(crate) struct ComponentPresence {
 
    component_id: ConnectorId,
 
    channels: Vec<LocalChannelPresence>,
 
}
 

	
 
#[derive(Debug, Clone)]
 
pub(crate) struct LocalChannelPresence {
 
    channel_id: ChannelId,
 
    is_closed: bool,
 
}
 

	
 
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
 
enum PresenceState {
 
    OnePresent, // one component reported the channel being open
 
    BothPresent, // two components reported the channel being open
 
    Closed, // one component reported the channel being closed
 
}
 

	
 
/// Record to hold channel state during the error-resolving mode of the leader.
 
/// This is used to determine when the sync region has grown to its largest
 
/// size. The structure is eventually consistent in the sense that a component
 
/// might initially presume a channel is open, only to figure out later it is
 
/// actually closed.
 
#[derive(Debug, Clone)]
 
struct ChannelPresence {
 
    added_by: ConnectorId,
 
    channel: ChannelId,
 
    both_sides_present: bool,
 
    owner_a: ConnectorId,
 
    owner_b: Option<ConnectorId>,
 
    id: ChannelId,
 
    state: PresenceState,
 
}
 

	
 
// TODO: Flatten? Flatten. Flatten everything.
 
#[derive(Debug)]
 
pub(crate) struct SolutionCombiner {
 
    local: Vec<ComponentLocalSolutions>, // used for finding solution
 
    presence: Vec<ChannelPresence>, // used to detect all channels present in case of failure
 
    failure_reported: bool,
 
}
 

	
 
struct CheckEntry {
 
    component_index: usize,         // component index in combiner's vector
 
    solution_index: usize,          // solution entry in the above component entry
 
    parent_entry_index: usize,      // parent that caused the creation of this checking entry
 
    match_index_in_parent: usize,   // index in the matches array of the parent
 
    solution_index_in_parent: usize,// index in the solution array of the match entry in the parent
 
}
 

	
 
enum LeaderConclusion {
 
    Solution(GlobalSolution),
 
    Failure,
 
}
 

	
 
impl SolutionCombiner {
 
    fn new() -> Self {
 
        return Self{
 
            local: Vec::new(),
 
            presence: Vec::new(),
 
            failure_reported: false,
 
        };
 
    }
 

	
 
    /// Adds a new local solution to the global solution storage. Will check the
 
    /// new local solutions for matching against already stored local solutions
 
    /// of peer connectors.
 
    fn add_solution_and_check_for_global_solution(&mut self, solution: LocalSolution) -> Option<GlobalSolution> {
 
        let component_id = solution.component;
 
        let solution = MatchedLocalSolution{
 
            final_branch_id: solution.final_branch_id,
 
            channel_mapping: solution.port_mapping,
 
            matches: Vec::new(),
 
        };
 

	
 
        // Create an entry for the solution for the particular component
 
        let component_exists = self.local.iter_mut()
 
            .enumerate()
 
            .find(|(_, v)| v.component == component_id);
 
        let (component_index, solution_index, new_component) = match component_exists {
 
@@ -994,115 +1071,132 @@ impl SolutionCombiner {
 

	
 
        // And now that we have the new solution-to-solution matches, we need to
 
        // add those in the appropriate storage.
 
        for new_component_match in new_component_matches {
 
            let other_component = &mut self.local[new_component_match.target_index];
 

	
 
            for other_solution_index in new_component_match.match_indices.iter().copied() {
 
                let other_solution = &mut other_component.solutions[other_solution_index];
 

	
 
                // Add a completely new entry for the component, or add it to
 
                // the existing component entry's matches
 
                match other_solution.matches.iter_mut()
 
                    .find(|v| v.target_id == component_id)
 
                {
 
                    Some(other_match) => {
 
                        other_match.match_indices.push(solution_index);
 
                    },
 
                    None => {
 
                        other_solution.matches.push(ComponentMatches{
 
                            target_id: component_id,
 
                            target_index: component_index,
 
                            match_indices: vec![solution_index],
 
                        })
 
                    }
 
                }
 
            }
 

	
 
            let cur_component = &mut self.local[component_index];
 
            let cur_solution = &mut cur_component.solutions[solution_index];
 

	
 
            match cur_solution.matches.iter_mut()
 
                .find(|v| v.target_id == new_component_match.target_id)
 
            {
 
                Some(other_match) => {
 
                    // Already have an entry
 
                    debug_assert_eq!(other_match.target_index, new_component_match.target_index);
 
                    other_match.match_indices.extend(&new_component_match.match_indices);
 
                },
 
                None => {
 
                    // Create a new entry
 
                    cur_solution.matches.push(new_component_match);
 
                }
 
            }
 
        }
 

	
 
        return self.check_for_global_solution(component_index, solution_index);
 
    }
 

	
 
    fn add_presence_and_check_for_global_failure(&mut self, present_component: ConnectorId, present: &[ChannelId]) -> bool {
 
        'new_channel_loop: for new_channel_id in present {
 
            for presence in &mut self.presence {
 
                if presence.channel == *new_channel_id {
 
                    if presence.added_by != present_component {
 
                        presence.both_sides_present = true;
 
    fn add_presence_and_check_for_global_failure(&mut self, component_id: ConnectorId, channels: &[LocalChannelPresence]) -> bool {
 
        'new_report_loop: for entry in channels {
 
            let mut found = false;
 

	
 
            for existing in &mut self.presence {
 
                if existing.id == entry.channel_id {
 
                    // Same entry. We only update if we have the second
 
                    // component coming in it owns one end of the channel, or if
 
                    // a component is telling us that the channel is (now)
 
                    // closed.
 
                    if entry.is_closed {
 
                        existing.state = PresenceState::Closed;
 
                    } else if component_id != existing.owner_a && existing.state != PresenceState::Closed {
 
                        existing.state = PresenceState::BothPresent;
 
                    }
 

	
 
                    continue 'new_channel_loop;
 
                    if existing.owner_a != component_id {
 
                        existing.owner_b = Some(component_id);
 
                    }
 

	
 
                    found = true;
 
                    break;
 
                }
 
            }
 

	
 
            // Not added yet
 
            if !found {
 
                self.presence.push(ChannelPresence{
 
                added_by: present_component,
 
                channel: *new_channel_id,
 
                both_sides_present: false,
 
                    owner_a: component_id,
 
                    owner_b: None,
 
                    id: entry.channel_id,
 
                    state: if entry.is_closed { PresenceState::Closed } else { PresenceState::OnePresent },
 
                });
 
            }
 
        }
 

	
 
        println!("DEBUGGERINO Presence is now:\n{:#?}", self.presence);
 

	
 
        return self.check_for_global_failure();
 
    }
 

	
 
    fn mark_failure_and_check_for_global_failure(&mut self) -> bool {
 
        self.failure_reported = true;
 
        return self.check_for_global_failure();
 
    }
 

	
 
    /// Checks if, starting at the provided local solution, a global solution
 
    /// can be formed.
 
    // TODO: At some point, check if divide and conquer is faster?
 
    fn check_for_global_solution(&self, initial_component_index: usize, initial_solution_index: usize) -> Option<GlobalSolution> {
 
        // Small trivial test necessary (but not sufficient) for a global
 
        // solution
 
        for component in &self.local {
 
            if !component.all_peers_present {
 
                return None;
 
            }
 
        }
 

	
 
        // Construct initial entry on stack
 
        let mut stack = Vec::with_capacity(self.local.len());
 
        stack.push(CheckEntry{
 
            component_index: initial_component_index,
 
            solution_index: initial_solution_index,
 
            parent_entry_index: 0,
 
            match_index_in_parent: 0,
 
            solution_index_in_parent: 0,
 
        });
 

	
 
        'check_last_stack: loop {
 
            let cur_index = stack.len() - 1;
 
            let cur_entry = &stack[cur_index];
 

	
 
            // Check if the current component is matching with all other entries
 
            let mut all_match = true;
 
            'check_against_existing: for prev_index in 0..cur_index {
 
                let prev_entry = &stack[prev_index];
 
                let prev_component = &self.local[prev_entry.component_index];
 
                let prev_solution = &prev_component.solutions[prev_entry.solution_index];
 

	
 
                for prev_matching_component in &prev_solution.matches {
 
                    if prev_matching_component.target_index == cur_entry.component_index {
 
                        // Previous entry has shared ports with the current
 
                        // entry, so see if we have a composable pair of
 
                        // solutions.
 
                        if !prev_matching_component.match_indices.contains(&cur_entry.solution_index) {
 
@@ -1204,160 +1298,188 @@ impl SolutionCombiner {
 

	
 
        // Just debugging here, TODO: @remove
 
        let mut total_num_channels = 0;
 
        for entry in &stack {
 
            let component = &self.local[entry.component_index];
 
            total_num_channels += component.solutions[0].channel_mapping.len();
 
        }
 

	
 
        total_num_channels /= 2;
 
        let mut final_mapping = Vec::with_capacity(total_num_channels);
 
        let mut total_num_checked = 0;
 

	
 
        for entry in &stack {
 
            let component = &self.local[entry.component_index];
 
            let solution = &component.solutions[entry.solution_index];
 

	
 
            for (channel_id, branch_id) in solution.channel_mapping.iter().copied() {
 
                match final_mapping.iter().find(|(v, _)| *v == channel_id) {
 
                    Some((_, encountered_branch_id)) => {
 
                        debug_assert_eq!(*encountered_branch_id, branch_id);
 
                        total_num_checked += 1;
 
                    },
 
                    None => {
 
                        final_mapping.push((channel_id, branch_id));
 
                    }
 
                }
 
            }
 
        }
 

	
 
        debug_assert_eq!(total_num_checked, total_num_channels);
 

	
 
        return Some(GlobalSolution{
 
            component_branches: final_branches,
 
            channel_mapping: final_mapping,
 
        });
 
    }
 

	
 
    /// Checks if all preconditions for global sync failure have been met
 
    fn check_for_global_failure(&self) -> bool {
 
        if !self.failure_reported {
 
            return false;
 
        }
 

	
 
        // Failure is reported, if all components are present then we may emit
 
        // the global failure broadcast
 
        // Check if all are present and we're preparing to fail this round
 
        let mut all_present = true;
 
        for presence in &self.presence {
 
            if !presence.both_sides_present {
 
            if presence.state == PresenceState::OnePresent {
 
                all_present = false;
 
                break;
 
            }
 
        }
 

	
 
        return all_present; // && failure_reported, which is checked above
 
    }
 

	
 
    /// Turns the entire (partially resolved) global solution into a structure
 
    /// that can be forwarded to a new parent. The new parent may then merge
 
    /// already obtained information.
 
    fn drain(&mut self) -> Option<SolutionCombiner> {
 
        if self.local.is_empty() && self.presence.is_empty() && !self.failure_reported {
 
            return None;
 
        }
 

	
 
        let result = SolutionCombiner{
 
            local: self.local.clone(),
 
            presence: self.presence.clone(),
 
            failure_reported: self.failure_reported,
 
        };
 

	
 
        self.local.clear();
 
        self.presence.clear();
 
        self.failure_reported = false;
 
        return Some(result);
 
    }
 

	
 
    // TODO: Entire routine is quite wasteful. Combine instead of doing all work
 
    //  again.
 
    fn combine(&mut self, combiner: SolutionCombiner) -> Option<LeaderConclusion> {
 
        self.failure_reported = self.failure_reported || combiner.failure_reported;
 

	
 
        // Handle local solutions
 
        if self.local.is_empty() {
 
            // Trivial case
 
            self.local = combiner.local;
 
        } else {
 
            for local in combiner.local {
 
                for matched in local.solutions {
 
                    let local_solution = LocalSolution{
 
                        component: local.component,
 
                        final_branch_id: matched.final_branch_id,
 
                        port_mapping: matched.channel_mapping,
 
                    };
 
                    let maybe_solution = self.add_solution_and_check_for_global_solution(local_solution);
 
                    if let Some(global_solution) = maybe_solution {
 
                        return Some(LeaderConclusion::Solution(global_solution));
 
                    }
 
                }
 
            }
 
        }
 

	
 
        // Handle channel presence
 
        if self.presence.is_empty() {
 
            // Trivial case
 
            self.presence = combiner.presence
 
        } else {
 
            for presence in combiner.presence {
 
                let global_failure = self.add_presence_and_check_for_global_failure(presence.added_by, &[presence.channel]);
 
                if global_failure {
 
                    return Some(LeaderConclusion::Failure);
 
                match self.presence.iter_mut().find(|v| v.id == presence.id) {
 
                    Some(entry) => {
 
                        // Combine entries. Take first that has Closed, then
 
                        // check first that has both, then check if they are
 
                        // combinable
 
                        if entry.state == PresenceState::Closed {
 
                            // Do nothing
 
                        } else if presence.state == PresenceState::Closed {
 
                            entry.owner_a = presence.owner_a;
 
                            entry.owner_b = presence.owner_b;
 
                            entry.state = PresenceState::Closed;
 
                        } else if entry.state == PresenceState::BothPresent {
 
                            // Again: do nothing
 
                        } else if presence.state == PresenceState::BothPresent {
 
                            entry.owner_a = presence.owner_a;
 
                            entry.owner_b = presence.owner_b;
 
                            entry.state = PresenceState::BothPresent;
 
                        } else {
 
                            // Both have one presence, combine into both present
 
                            debug_assert!(entry.state == PresenceState::OnePresent && presence.state == PresenceState::OnePresent);
 
                            entry.owner_b = Some(presence.owner_a);
 
                        }
 
                    },
 
                    None => {
 
                        self.presence.push(presence);
 
                    }
 
                }
 
            }
 

	
 
            // After adding everything we might have immediately found a solution
 
            if self.check_for_global_failure() {
 
                return Some(LeaderConclusion::Failure);
 
            }
 
        }
 

	
 
        return None;
 
    }
 

	
 
    fn clear(&mut self) {
 
        self.local.clear();
 
        self.presence.clear();
 
        self.failure_reported = false;
 
    }
 
}
 

	
 
// -----------------------------------------------------------------------------
 
// Generic Helpers
 
// -----------------------------------------------------------------------------
 

	
 
/// Recursively goes through the value group, attempting to find ports.
 
/// Duplicates will only be added once.
 
pub(crate) fn find_ports_in_value_group(value_group: &ValueGroup, ports: &mut Vec<PortIdLocal>) {
 
    // Helper to check a value for a port and recurse if needed.
 
    use crate::protocol::eval::Value;
 

	
 
    fn find_port_in_value(group: &ValueGroup, value: &Value, ports: &mut Vec<PortIdLocal>) {
 
        match value {
 
            Value::Input(port_id) | Value::Output(port_id) => {
 
                // This is an actual port
 
                let cur_port = PortIdLocal::new(port_id.0.u32_suffix);
 
                for prev_port in ports.iter() {
 
                    if *prev_port == cur_port {
 
                        // Already added
 
                        return;
 
                    }
 
                }
 

	
 
                ports.push(cur_port);
 
            },
 
            Value::Array(heap_pos) |
 
            Value::Message(heap_pos) |
 
            Value::String(heap_pos) |
 
            Value::Struct(heap_pos) |
 
            Value::Union(_, heap_pos) => {
 
                // Reference to some dynamic thing which might contain ports,
 
                // so recurse
 
                let heap_region = &group.regions[*heap_pos as usize];
 
                for embedded_value in heap_region {
 
                    find_port_in_value(group, embedded_value, ports);
 
                }
src/runtime2/inbox.rs
Show inline comments
 
use std::sync::Mutex;
 
use std::collections::VecDeque;
 

	
 
use crate::protocol::eval::ValueGroup;
 
use crate::runtime2::consensus::SolutionCombiner;
 
use crate::runtime2::consensus::{ComponentPresence, SolutionCombiner};
 
use crate::runtime2::port::ChannelId;
 

	
 
use super::ConnectorId;
 
use super::branch::BranchId;
 
use super::consensus::{GlobalSolution, LocalSolution};
 
use super::port::PortIdLocal;
 

	
 
// TODO: Remove Debug derive from all types
 

	
 
#[derive(Debug, Copy, Clone)]
 
pub(crate) struct ChannelAnnotation {
 
    pub channel_id: ChannelId,
 
    pub registered_id: Option<BranchMarker>,
 
    pub expected_firing: Option<bool>,
 
}
 

	
 
/// Marker for a branch in a port mapping. A marker is, like a branch ID, a
 
/// unique identifier for a branch, but differs in that a branch only has one
 
/// branch ID, but might have multiple associated markers (i.e. one branch
 
/// performing a `put` three times will generate three markers.
 
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
 
pub(crate) struct BranchMarker{
 
    marker: u32,
 
}
 

	
 
impl BranchMarker {
 
    #[inline]
 
    pub(crate) fn new(marker: u32) -> Self {
 
        debug_assert!(marker != 0);
 
        return Self{ marker };
 
    }
 

	
 
    #[inline]
 
    pub(crate) fn new_invalid() -> Self {
 
        return Self{ marker: 0 }
 
    }
 
}
 

	
 
/// The header added by the synchronization algorithm to all.
 
#[derive(Debug, Clone)]
 
pub(crate) struct SyncHeader {
 
    pub sending_component_id: ConnectorId,
 
    pub highest_component_id: ConnectorId,
 
    pub sync_round: u32,
 
}
 

	
 
/// The header added to data messages
 
#[derive(Debug, Clone)]
 
pub(crate) struct DataHeader {
 
    pub expected_mapping: Vec<ChannelAnnotation>,
 
    pub sending_port: PortIdLocal,
 
    pub target_port: PortIdLocal,
 
    pub new_mapping: BranchMarker,
 
}
 

	
 
/// A data message is a message that is intended for the receiver's PDL code,
 
/// but will also be handled by the consensus algorithm
 
#[derive(Debug, Clone)]
 
pub(crate) struct DataMessage {
 
    pub sync_header: SyncHeader,
 
    pub data_header: DataHeader,
 
    pub content: ValueGroup,
 
}
 

	
 
#[derive(Debug)]
 
pub(crate) enum SyncCompContent {
 
    LocalFailure, // notifying leader that component has failed (e.g. timeout, whatever)
 
    LocalSolution(LocalSolution), // sending a local solution to the leader
 
    PartialSolution(SolutionCombiner), // when new leader is detected, forward all local results
 
    GlobalSolution(GlobalSolution), // broadcasting to everyone
 
    GlobalFailure, // broadcasting to everyone
 
    AckFailure, // acknowledgement of failure to leader
 
    Notification, // just a notification (so purpose of message is to send the SyncHeader)
 
    Presence(ConnectorId, Vec<ChannelId>), // notifying leader of component presence (needed to ensure failing a round involves all components in a sync round)
 
    Presence(ComponentPresence), // notifying leader of component presence (needed to ensure failing a round involves all components in a sync round)
 
}
 

	
 
/// A sync message is a message that is intended only for the consensus
 
/// algorithm. The message goes directly to a component.
 
#[derive(Debug)]
 
pub(crate) struct SyncCompMessage {
 
    pub sync_header: SyncHeader,
 
    pub target_component_id: ConnectorId,
 
    pub content: SyncCompContent,
 
}
 

	
 
#[derive(Debug)]
 
pub(crate) enum SyncPortContent {
 
    SilentPortNotification,
 
    NotificationWave,
 
}
 

	
 
/// A sync message intended for the consensus algorithm. This message does not
 
/// go to a component, but through a channel (and results in potential
 
/// rerouting) because we're not sure about the ID of the component that holds
 
/// the other end of the channel.
 
#[derive(Debug)]
 
pub(crate) struct SyncPortMessage {
 
    pub sync_header: SyncHeader,
 
    pub source_port: PortIdLocal,
 
    pub target_port: PortIdLocal,
 
    pub content: SyncPortContent,
 
}
 

	
 
#[derive(Debug)]
 
pub(crate) enum SyncControlContent {
 
    ChannelIsClosed(PortIdLocal), // contains port that is owned by the recipient of the message
 
}
 

	
 
/// A sync control message: originating from the scheduler, but intended for the
 
/// current sync round of the recipient. Every kind of consensus algorithm must
 
/// be able to handle such a message.
 
#[derive(Debug)]
 
pub(crate) struct SyncControlMessage {
 
    // For now these control messages are only aimed at components. Might change
 
    // in the future. But for now we respond to messages from components that
 
    // have, because of that message, published their ID.
 
    pub in_response_to_sync_round: u32,
 
    pub target_component_id: ConnectorId,
 
    pub content: SyncControlContent,
 
}
 

	
 
/// A control message is a message intended for the scheduler that is executing
src/runtime2/tests/mod.rs
Show inline comments
 
mod network_shapes;
 
mod api_component;
 
mod speculation;
 
mod data_transmission;
 
mod sync_failure;
 

	
 
use super::*;
 
use crate::{PortId, ProtocolDescription};
 
use crate::common::Id;
 
use crate::protocol::eval::*;
 
use crate::runtime2::native::{ApplicationSyncAction};
 

	
 
// Generic testing constants, use when appropriate to simplify stress-testing
 
// pub(crate) const NUM_THREADS: u32 = 3;     // number of threads in runtime
 
// pub(crate) const NUM_INSTANCES: u32 = 7;   // number of test instances constructed
 
// pub(crate) const NUM_LOOPS: u32 = 8;       // number of loops within a single test (not used by all tests)
 

	
 
pub(crate) const NUM_THREADS: u32 = 2;
 
pub(crate) const NUM_THREADS: u32 = 4;
 
pub(crate) const NUM_INSTANCES: u32 = 1;
 
pub(crate) const NUM_LOOPS: u32 = 1;
 

	
 

	
 
fn create_runtime(pdl: &str) -> Runtime {
 
    let protocol = ProtocolDescription::parse(pdl.as_bytes()).expect("parse pdl");
 
    let runtime = Runtime::new(NUM_THREADS, protocol);
 

	
 
    return runtime;
 
}
 

	
 
fn run_test_in_runtime<F: Fn(&mut ApplicationInterface)>(pdl: &str, constructor: F) {
 
    let protocol = ProtocolDescription::parse(pdl.as_bytes())
 
        .expect("parse PDL");
 
    let runtime = Runtime::new(NUM_THREADS, protocol);
 

	
 
    let mut api = runtime.create_interface();
 
    for _ in 0..NUM_INSTANCES {
 
        constructor(&mut api);
 
    }
 
}
 

	
 
pub(crate) struct TestTimer {
 
    name: &'static str,
 
    started: std::time::Instant
 
}
 

	
 
impl TestTimer {
 
    pub(crate) fn new(name: &'static str) -> Self {
 
        Self{ name, started: std::time::Instant::now() }
 
    }
 
}
 

	
 
impl Drop for TestTimer {
 
    fn drop(&mut self) {
 
        let delta = std::time::Instant::now() - self.started;
 
        let nanos = (delta.as_secs_f64() * 1_000_000.0) as u64;
 
        let millis = nanos / 1000;
 
        let nanos = nanos % 1000;
 
        println!("[{}] Took {:>4}.{:03} ms", self.name, millis, nanos);
 
    }
 
}
src/runtime2/tests/sync_failure.rs
Show inline comments
 
@@ -20,56 +20,56 @@ fn test_local_sync_failure() {
 
        u32[] only_allows_index_0 = { 1 };
 
        auto never_gonna_get = only_allows_index_0[1];
 
        while (true) sync {}
 
    }
 
    ";
 

	
 
    // let thing = TestTimer::new("local_sync_failure");
 
    run_test_in_runtime(CODE, |api| {
 
        api.create_connector("", "immediate_failure_outside_sync", ValueGroup::new_stack(Vec::new()))
 
            .expect("create component");
 

	
 
        api.create_connector("", "immediate_failure_inside_sync", ValueGroup::new_stack(Vec::new()))
 
            .expect("create component");
 
    })
 
}
 

	
 
#[test]
 
fn test_shared_sync_failure() {
 
    // Same as above. One of the components should fail, the other should follow
 
    // suit because it cannot complete a sync round. We intentionally have an
 
    // infinite loop in the while condition because we need at least two loops
 
    // for the last error to get picked up.
 
    const CODE: &'static str = "
 
    enum Location { BeforeSync, AfterPut, AfterGet, AfterSync, Never }
 
    primitive failing_at_location(in<bool> input, out<bool> output, Location loc) {
 
        u32[] failure_array = {};
 
        while (true) {
 
            if (loc == Location::BeforeSync) failure_array[0];
 
            sync {
 
                put(output, true);
 
                if (loc == Location::AfterPut) failure_array[0];
 
                auto received = get(input);
 
                assert(received);
 
                if (loc == Location::AfterGet) failure_array[0];
 
            }
 
            if (loc == Location::AfterSync) failure_array[0];
 
        }
 
    }
 

	
 
    composite constructor(Location loc) {
 
        channel output_a -> input_a;
 
        channel output_b -> input_b;
 
        new failing_at_location(input_a, output_b, Location::Never);
 
        new failing_at_location(input_b, output_a, loc);
 
    }
 
    ";
 

	
 
    run_test_in_runtime(CODE, |api| {
 
        for variant in 3..4 { // all `Location` enum variants, except `Never`.
 
        for variant in 0..4 { // all `Location` enum variants, except `Never`.
 
            // Create the channels
 
            api.create_connector("", "constructor", ValueGroup::new_stack(vec![
 
                Value::Enum(variant)
 
            ])).expect("create connector");
 
        }
 
    })
 
}
 
\ No newline at end of file
0 comments (0 inline, 0 general)