Yehowshua/ZIm
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Use an unbounded channel for peer's outgoing messages

Browse source 
Using a bounded channel may have blocked the collaboration server
from making progress handling RPC traffic.

There's no need to apply backpressure to calling code within the
same process - suspending a task that is attempting to call `send` has
an even greater memory cost than just buffering a protobuf message.

We do still want a bounded channel for incoming messages, so that
we provide backpressure to noisy peers - blocking their writes as opposed
to allowing them to buffer arbitrarily many messages in our server.

Co-Authored-By: Antonio Scandurra <me@as-cii.com>
Co-Authored-By: Nathan Sobo <nathan@zed.dev>
This commit is contained in:
Max Brunsfeld 2022-02-07 12:27:13 -08:00
parent 82afacd33d
commit d4fe1115e7
7 changed files with 341 additions and 472 deletions
Download patch file Download diff file Expand all files Collapse all files

6
crates/client/src/test.rs
View file

@ -118,8 +118,8 @@ impl FakeServer {
self.forbid_connections.store(false, SeqCst);
}
pub async fn send<T: proto::EnvelopedMessage>(&self, message: T) {
self.peer.send(self.connection_id(), message).await.unwrap();
pub fn send<T: proto::EnvelopedMessage>(&self, message: T) {
self.peer.send(self.connection_id(), message).unwrap();
}
pub async fn receive<M: proto::EnvelopedMessage>(&self) -> Result<TypedEnvelope<M>> {
@ -148,7 +148,7 @@ impl FakeServer {
receipt: Receipt<T>,
response: T::Response,
) {
self.peer.respond(receipt, response).await.unwrap()
self.peer.respond(receipt, response).unwrap()
}
fn connection_id(&self) -> ConnectionId {