perf: request path — XRPC transport, serialization, auth and server allocations, plus debloat #194

New issue

Closed

opened 2026-10-05 02:10:26 +00:00 by Grandiras · 0 comments

Grandiras commented

2026-10-05 02:10:26 +00:00

Owner

Goal

Cut allocations and latency on the per-request path, and reduce code lines, without changing behaviour.

Scope

src/ATProtoNet/{Http,Serialization,Auth,Caching,Models}, AtProtoClient.cs, RecordCollection.cs, and src/ATProtoNet.Server outside Spaces/.

Measured before/after with BenchmarkDotNet (allocated bytes and time per operation); duplication and dead code removed along the way.

### Goal Cut allocations and latency on the per-request path, and reduce code lines, without changing behaviour. ### Scope `src/ATProtoNet/{Http,Serialization,Auth,Caching,Models}`, `AtProtoClient.cs`, `RecordCollection.cs`, and `src/ATProtoNet.Server` outside `Spaces/`. Measured before/after with BenchmarkDotNet (allocated bytes and time per operation); duplication and dead code removed along the way.

Grandiras referenced this issue from a commit

2026-10-05 02:29:45 +00:00

perf: leaner XRPC request path (HTTP/2, Content-Length bodies, cheaper DPoP proofs) and dead-code removal (closes #194)

Grandiras referenced this issue from a pull request that will close it,

2026-10-05 02:29:46 +00:00

perf: leaner XRPC request path (HTTP/2, single JSON serialization, DPoP) #201

Grandiras referenced this issue from a commit

2026-10-05 02:58:16 +00:00

perf: leaner XRPC request path (HTTP/2, Content-Length bodies, cheaper DPoP proofs) and dead-code removal (closes #194)

Grandiras referenced this issue from a commit

2026-10-05 04:07:21 +00:00

perf: leaner XRPC request path (HTTP/2, Content-Length bodies, cheaper DPoP proofs) and dead-code removal (closes #194)