Fix: keep a newer same-run TensorMap entry when its predecessor is consumed

[ChaoWao]

Fixes #1480

Problem

TensorMap::erase_task_outputs(run_id, keys) erased every key unconditionally.
Orchestrator::on_consumed calls it with the consumed task's output_keys, so
the cleanup could drop a mapping that a later task in the same run already
owns.

Two OUTPUT / OUTPUT_EXISTING writers of one key carry no edge between them —
that path inserts without looking up — so the first can reach CONSUMED while
the second is still running:

1. A: OUTPUT K → insert(run, K, A)
2. B: OUTPUT K → insert(run, K, B) (overwrites; A and B unordered)
3. A consumed → erase_task_outputs(run, [K]) removes the entry pointing at B
4. C: INPUT K → lookup returns INVALID_SLOT → no dependency on B

C can then be dispatched and read K before B has written it.

INOUT is unaffected: it looks up before inserting, so B depends on A and A
cannot reach CONSUMED until B has released its reference.

Fix

erase_task_outputs takes the consumed slot and drops a key only while it still
maps to that slot. This mirrors how the L2 TensorMap reclaims a retired task's
entries — pto_tensormap.h::cleanup_retired, "Only remove if this entry belongs
to the retiring task (slot may have been reused by a newer task)".

Tests

• OrchestratorFixture.ConsumingASupersededProducerKeepsTheNewerMapping — the
  end-to-end repro. Verified failing on the pre-fix code: C came back with
  fanin_producers.size() == 0 and fanin_count == 0.
• TensorMap.EraseKeepsNewerSameRunProducer — same-run WAW at the map level,
  alongside the existing cross-run isolation test. Also asserts the owning
  slot's erase still removes the entry.

Existing erase_task_outputs call sites in test_tensormap.cpp take the owner
slot argument.

Verification

• ctest --test-dir tests/ut/cpp/build -LE requires_hardware — 60/60 passed
• pytest tests/ut/py — 824 passed, 2 skipped
• pytest tests/st/worker --platform a2a3sim --device 0-3 (level-3 collectives,
  the hierarchical orchestrator's ST coverage) — 18/18 passed
• pre-commit run on the touched files — all hooks pass

Host-only change; no hardware run required.

Left open

Whether two unordered OUTPUT writers of one key should be a diagnosed error
rather than a silently unordered WAW — raised in the issue, deliberately not
decided here.

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

@ChaoWao, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 28 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 488e2837-d69c-4c8f-9f9d-7ca34f771e2f

📥 Commits

Reviewing files that changed from the base of the PR and between e568310 and d0d4a9b.

📒 Files selected for processing (6)

• docs/orchestrator.md
• src/common/hierarchical/orchestrator.cpp
• src/common/hierarchical/tensormap.cpp
• src/common/hierarchical/tensormap.h
• tests/ut/cpp/hierarchical/test_orchestrator.cpp
• tests/ut/cpp/hierarchical/test_tensormap.cpp

---

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.}