Allow to broadcast network messages in parallel #7542
Conversation
vstakhov
added
I10-optimisation
|
@paritytech-cicd-pr Requester could not be detected as a member of an allowed organization. |
| } | ||
|
|
||
| let _: Vec<()> = delayed_messages.collect().await; |
There was a problem hiding this comment.
technically the bridge is still stopped until all futures complete here.
There was a problem hiding this comment.
Yes, but other subsystems have a chance to receive updates if some other subsystem is slow. That's the goal of this PR technically.
|
bot rebase |
|
Error: Command 'Command { std: cd "/storage/repositories/polkadot" && "git" "merge" "origin/master" "--no-ff" "--no-edit", kill_on_drop: false }' failed with status Some(1); output: no output |
|
The CI pipeline was cancelled due to failure one of the required jobs. |
|
bot rebase |
…idge-small-refactor
|
Rebased |
|
Versi burning has not revealed any significant issues. However, we might want to burn it for a longer time with extreme disputes load and compare the metrics before/after. What do you think @sandreim ? |
| sender.send_messages(event.focus().map(BitfieldDistributionMessage::from)).await; | ||
| sender.send_messages(event.focus().map(ApprovalDistributionMessage::from)).await; | ||
| sender.send_messages(event.focus().map(GossipSupportMessage::from)).await; | ||
| if let Ok(msg) = event.focus().map(StatementDistributionMessage::from) { |
There was a problem hiding this comment.
Is it ok to ignore a possible error?
This PR addresses multiple issues pending:
This PR needs Versi burning as it involves some fundamental node components change.