Ratify M10.1 timing and workload budgets with session traces
Foundation / checks (deploy) (push) Successful in 1m0s
Foundation / checks (ecs) (push) Successful in 3m34s
Foundation / checks (network) (push) Successful in 8m8s
Foundation / checks (client) (push) Successful in 14m49s
Foundation / checks (tsan) (push) Successful in 7m35s
Foundation / postgres-tools (push) Canceled after 0s
Foundation / server-image (push) Canceled after 0s
Foundation / image (push) Canceled after 0s
Foundation / checks (postgres) (push) Canceled after 16m18s
Foundation / checks (checks) (push) Canceled after 29m27s

ADR-0089 records the maintainer-ratified 50 ms tick, 1 s flush with final
outcomes p95 <= 1.5 s, responsiveness targets and reference hardware, plus
derived scheduling, sequence, identity, baseline, queue/byte/work budgets and the
exact/tolerant determinism contract (schemas/network/m10-budgets-v1.md). Adds
SessionClient traffic counters, `evo-dev-session trace` and a reproducible trace
script; development-reference results are in M10_TIMING_TRACES. Records that
genome v1 and metabolism rates are per tick, so the quantum fixes biology rates.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
dingo
2026-10-03 12:29:58 -04:00
co-authored by Claude Opus 5.5
parent 6152d58fa6
commit 819137c1df
15 changed files with 557 additions and 12 deletions
+8
View File
@@ -50,6 +50,13 @@ struct ClientEvent {
IntentOutcome outcome{};
std::uint64_t sequence{};
};
// Received-frame counters for timing and bandwidth experiments (M10.1); bytes include
// the frame header. Monotonic for the client's lifetime, across reconnects.
struct ClientTraffic {
std::uint64_t frames{}, bytes{};
std::uint64_t environment_views{}, environment_view_bytes{};
std::uint64_t organism_views{}, organism_view_bytes{};
};
enum class SessionClientError : std::uint8_t { configuration };
class SessionClient {
public:
@@ -73,6 +80,7 @@ class SessionClient {
// Attempts that failed before any session event (TCP connect or TLS handshake), so a
// caller can bound "unreachable" without an event per retry. Never reset.
[[nodiscard]] std::uint64_t unreachable_attempts() const noexcept;
[[nodiscard]] ClientTraffic traffic() const noexcept;
// Drains at most max_events and resumes reconnect after local backpressure.
[[nodiscard]] std::vector<ClientEvent> take_events();