Skip to content

Remove broadcast channel#317

Open
Shourya742 wants to merge 7 commits intostratum-mining:mainfrom
Shourya742:2026-03-04-remove-broadcast
Open

Remove broadcast channel#317
Shourya742 wants to merge 7 commits intostratum-mining:mainfrom
Shourya742:2026-03-04-remove-broadcast

Conversation

@Shourya742
Copy link
Collaborator

@Shourya742 Shourya742 commented Mar 4, 2026

closes: #219 #220 #221

@Shourya742 Shourya742 force-pushed the 2026-03-04-remove-broadcast branch 2 times, most recently from 679b268 to acc10b6 Compare March 4, 2026 18:52
@Shourya742 Shourya742 force-pushed the 2026-03-04-remove-broadcast branch 2 times, most recently from 8333bc8 to 39d52cd Compare March 14, 2026 08:46
@Shourya742 Shourya742 marked this pull request as ready for review March 14, 2026 11:35
@Shourya742 Shourya742 force-pushed the 2026-03-04-remove-broadcast branch 2 times, most recently from 557e6d0 to 6f6464c Compare March 15, 2026 10:00
This was linked to issues Mar 16, 2026
replace tokio::sync::broadcast for Pool's ChannelManager to Downstream sender #220
Open
replace tokio::sync::broadcast on JDC's ChannelManager to Downstream sender #221
Open
@plebhash
Copy link
Member

plebhash commented Mar 24, 2026

it seems we would need to complete the DashMap migration to continue on the current approach here

we should keep the scope here focused on removal of broadcast channels only, without trying to optimize orthogonal aspects (as much as we need maps for the solution here)

IMO we should follow the current lock/map patterns we already have on JDC/Pool/tProxy, and then carefully evaluate the DashMap migration later

@Shourya742 Shourya742 force-pushed the 2026-03-04-remove-broadcast branch from 825b901 to 73f7bd6 Compare March 25, 2026 10:13
@Shourya742
Copy link
Collaborator Author

Shourya742 commented Mar 25, 2026

it seems we would need to complete the DashMap migration to continue on the current approach here

we should keep the scope here focused on removal of broadcast channels only, without trying to optimize orthogonal aspects (as much as we need maps for the solution here)

IMO we should follow the current lock/map patterns we already have on JDC/Pool/tProxy, and then carefully evaluate the DashMap migration later

Adapted the PR, removed dashmap with Mutex for now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@plebhash plebhash plebhash left review comments

At least 1 approving review is required to merge this pull request.

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

2 participants

@Shourya742 @plebhash