perf(message): reduce hot-path allocations

[jlucaso1]

Summary

• process unobserved ACK and receipt nodes inline and bypass hot async_trait allocations
• batch in-memory Signal session writes and remove temporary lookup and message-secret allocations
• preserve concurrent dispatch for observed events and synchronous receipt ACKs

Impact

DHAT ping-pong allocation cost per cycle dropped from 46,088 to 40,542 bytes (-12.0%) and from 254.84 to 231.60 blocks (-9.1%).

Validation

• cargo test -p wacore --lib
• cargo test -p whatsapp-rust --lib
• cargo check -p whatsapp-rust --lib --features tracing
• cargo clippy -p whatsapp-rust -p wacore --tests -- -D warnings
• direct, group, and receipt-observer benchmark smoke tests

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository UI (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: f9d440c0-430b-41da-8efc-1bcf3dd1eb71

📥 Commits

Reviewing files that changed from the base of the PR and between 3a6ef95 and 65aaae0.

📒 Files selected for processing (7)

• src/client/device_registry.rs
• src/client/node_io.rs
• src/handlers/message.rs
• src/msg_secret_buffer.rs
• src/receipt.rs
• src/send/mod.rs
• wacore/src/store/in_memory.rs

---

📝 Walkthrough

Summary by CodeRabbit

• Improvements
  • Improved responsiveness when processing messages, acknowledgements, and delivery receipts.
  • Enhanced message handling reliability during high-volume or backlogged conditions.
  • Improved device lookup consistency across alternate identifiers and account changes.
  • Streamlined secure message buffering, including more efficient single-message processing.
  • Improved session updates when multiple records are saved together.
• Bug Fixes
  • Improved cache invalidation so device-related updates are reflected more consistently.

Walkthrough

The changes optimize device lookup iteration, route selected stanzas through inline handlers, add single-entry message-secret buffering, and implement batched in-memory session updates with tests.

Changes

Runtime path and storage updates

Layer / File(s)	Summary
Device lookup iteration and alias tracking src/client/device_registry.rs	Lookup keys are produced and consumed as iterators, while device-cache alias tracking now uses lookup.all_keys().
Inline stanza and acknowledgement dispatch src/client/node_io.rs, src/handlers/message.rs, src/receipt.rs	Acknowledgements, receipts, and messages gain direct inline handling paths with async wrappers retained for trait and call-site compatibility.
Single-entry secret buffering src/msg_secret_buffer.rs, src/send/mod.rs	Pending-entry insertion is centralized, and outbound secret persistence uses the new single-entry queue method.
Batch session updates and validation wacore/src/store/in_memory.rs	The in-memory store updates session batches under one lock and tests insertion, replacement, and retention of other entries.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
  participant MessageLoop
  participant NodeProcessor
  participant InlineHandler
  participant ChatLane
  MessageLoop->>NodeProcessor: process decrypted node
  NodeProcessor->>InlineHandler: handle ack, receipt, or message inline
  InlineHandler->>ChatLane: enqueue message for chat
  ChatLane-->>InlineHandler: enqueue result
  InlineHandler-->>NodeProcessor: handled status and cancellation

Loading

Possibly related PRs

• oxidezap/whatsapp-rust#978: Modifies the same ACK handling and response-waiter dispatch path.
• oxidezap/whatsapp-rust#831: Changes the same buffered message-secret persistence path.
• oxidezap/whatsapp-rust#824: Changes device-list cache invalidation and touched-alias behavior.

Suggested labels: performance

Suggested reviewers: cubic-dev-ai, greptile-apps

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly matches the PR’s main goal of reducing hot-path allocations in message handling.
Description check	✅ Passed	The description is directly related to the changeset and accurately summarizes the allocation-focused refactors.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch agent/reduce-message-path-allocations

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[greptile-apps]

Greptile Summary

This PR reduces allocations on message and receipt hot paths. The main changes are:

• Inline handling for unobserved ack, receipt, and message nodes.
• Borrowed lookup-key iteration in the device registry.
• Single-entry message-secret buffering for outbound sends.
• Batched in-memory Signal session writes.

Confidence Score: 5/5

This looks safe to merge.

• No blocking issues found in the changed code.
• The inline handler paths keep the same handler bodies and preserve observer-aware dispatch.
• The buffer and batch changes keep the existing write and lookup semantics.

Important Files Changed

Filename	Overview
src/client/node_io.rs	Adds inline fast paths for built-in node handling while keeping observer and waiter gates.
src/handlers/message.rs	Extracts the message handler body into a reusable inline method.
src/receipt.rs	Moves receipt handling into a synchronous inline helper with the async wrapper preserved.
src/msg_secret_buffer.rs	Refactors pending insertion and scheduling, and adds a single-entry queue path.
src/send/mod.rs	Uses the single-entry message-secret queue path for outbound secret persistence.
src/client/device_registry.rs	Replaces temporary lookup-key vectors with borrowed iterators.
wacore/src/store/in_memory.rs	Adds a batched in-memory Signal session write implementation.

_{Reviews (1): Last reviewed commit: "perf(message): reduce hot-path allocatio..." | Re-trigger Greptile}

[cubic-dev-ai]

All reported issues were addressed across 7 files

Confidence score: 5/5

• Safe to merge after the addressed issues were fixed.

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>

[github-actions]

📦 Binary size report

Metric	main	PR	Δ
bin size (stripped)	10.83 MiB	10.83 MiB	+1.97 KiB (+0.02%) 🔺
bin .text	8.83 MiB	8.83 MiB	+1.62 KiB (+0.02%) 🔺
bin allocated (text+data+bss)	10.83 MiB	10.83 MiB	+104 B (+0.00%) 🔺
llvm-lines wacore	505,440	505,783	+343 (+0.07%) 🔺
llvm-lines wacore copies	17,345	17,358	+13 (+0.07%) 🔺
llvm-lines whatsapp-rust lib	767,242	767,999	+757 (+0.10%) 🔺
llvm-lines whatsapp-rust lib copies	24,947	24,916	-31 (-0.12%) 🔽
deps crates (Cargo.lock)	472	472	0

.text per crate

Crate	main	PR	Δ
.text whatsapp_rust	1.65 MiB	1.65 MiB	+2.31 KiB (+0.14%) 🔺
.text wacore	526.09 KiB	526.43 KiB	+348 B (+0.06%) 🔺
.text wacore_binary	157.88 KiB	157.88 KiB	0
.text wacore_libsignal	179.42 KiB	179.42 KiB	0
.text wacore_appstate	158.25 KiB	158.25 KiB	0
.text wacore_noise	26.05 KiB	26.05 KiB	0
.text waproto	1.60 MiB	1.60 MiB	0
.text whatsapp_rust_sqlite_storage	513.00 KiB	512.90 KiB	-98 B (-0.02%) 🔽
.text whatsapp_rust_tokio_transport	43.79 KiB	43.89 KiB	+98 B (+0.22%) 🔺
.text whatsapp_rust_ureq_http_client	10.47 KiB	10.47 KiB	0
.text std	1.00 MiB	1.00 MiB	-1.47 KiB (-0.14%) 🔽
.text other deps	2.95 MiB	2.95 MiB	+428 B (+0.01%) 🔺

Top movers (cargo-bloat attribution)

Crate	main	PR	Δ
whatsapp_rust	1.65 MiB	1.65 MiB	+2.31 KiB (+0.14%)
std	1.00 MiB	1.00 MiB	-1.47 KiB (-0.14%)

Baseline: 3a6ef95ee (latest main run) · Head: 79683c668 · Graphs