Triage pipeline could not create a sub-issue for auto-decomposition (API error or repo permissions). Marking this blocked:needs-decision to prevent re-evaluation. Remove the label to retry.

## Triage: decompose failed Triage pipeline could not create a sub-issue for auto-decomposition (API error or repo permissions). Marking this `blocked:needs-decision` to prevent re-evaluation. Remove the label to retry.

Mark added the

blocked:preflight

label

2026-07-31 04:26:45 +00:00

Mark removed the

blocked:needs-decision

blocked:preflight

labels

2026-08-11 04:33:15 +00:00

Mark commented

2026-08-12 08:07:28 +00:00

Author

Owner

Fixup closure: deliverable already on master

GPU-001 (GPU Detection and Allocation) is implemented and verified on origin/master:

GPU detection: primitives/_bone/gpu_detector.py — detect_gpu() queries nvidia-smi (name, memory, driver, uuid) and returns device list/count. Present since commit 16315ee (feat: Implement Phase 3.2 GPU Capability Routing, 2026-03-27, predates this issue).
GPU allocation: runtime/gpu_manager.py — GPUDevice/GPUNode/GPUManager with allocation helpers. Present since the Phase 0-4/Phase 3.2 implementation commits (2026-03-14 through 2026-03-31).
Extended by the sibling cards: multi-GPU support (#196, commit 39edeaf) and GPU monitoring (#197, commit f98427f), both closed.

Verification: pytest tests/unit/test_gpu_manager.py tests/unit/test_multi_gpu_routing.py tests/unit/test_gpu_monitor.py → 56 passed.

Why not --resume: this issue has no body/spec and no #195-specific plan or commit; the June salvage-state entry claims phases complete but produced no #195 commit (stale residue — the runner controller stalled at Review, same pattern as mesh-seed #78/#199). A live kilo-gsd run for #202 is in progress on the shared worktree; resuming #195 would collide with it. Closing with evidence instead.

## Fixup closure: deliverable already on master GPU-001 (GPU Detection and Allocation) is implemented and verified on `origin/master`: - **GPU detection**: `primitives/_bone/gpu_detector.py` — `detect_gpu()` queries nvidia-smi (name, memory, driver, uuid) and returns device list/count. Present since commit 16315ee (feat: Implement Phase 3.2 GPU Capability Routing, 2026-03-27, predates this issue). - **GPU allocation**: `runtime/gpu_manager.py` — GPUDevice/GPUNode/GPUManager with allocation helpers. Present since the Phase 0-4/Phase 3.2 implementation commits (2026-03-14 through 2026-03-31). - Extended by the sibling cards: multi-GPU support (#196, commit 39edeaf) and GPU monitoring (#197, commit f98427f), both closed. **Verification**: `pytest tests/unit/test_gpu_manager.py tests/unit/test_multi_gpu_routing.py tests/unit/test_gpu_monitor.py` → 56 passed. **Why not --resume**: this issue has no body/spec and no #195-specific plan or commit; the June salvage-state entry claims phases complete but produced no #195 commit (stale residue — the runner controller stalled at Review, same pattern as mesh-seed #78/#199). A live kilo-gsd run for #202 is in progress on the shared worktree; resuming #195 would collide with it. Closing with evidence instead.

Mark closed this issue

2026-08-12 08:07:49 +00:00

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

Mark/mesh-seed#195

No description provided.

Rows
Columns