Skip to content

feat(dcu-kernel-auto-opt): harness planner, gate policy and DKAO tooling - #34

Open
fms-zth wants to merge 1 commit into
HuangPuStar:next-appfrom
fms-zth:feat/dcu-kernel-auto-opt-20260911
Open

feat(dcu-kernel-auto-opt): harness planner, gate policy and DKAO tooling#34
fms-zth wants to merge 1 commit into
HuangPuStar:next-appfrom
fms-zth:feat/dcu-kernel-auto-opt-20260911

Conversation

@fms-zth

@fms-zth fms-zth commented Sep 11, 2026

Copy link
Copy Markdown
Contributor

Content imported from the DCU dev container tree (/root/zth_agent/MetaInfer), 3-way merged onto the merged PR #26 head so the next-app alignment and CI fixes stay in place. Scope is limited to this task plugin.

Added:

  • orchestrator: planner.py (+ planner_policy/manifest/gates), gate_policy.py, gpu_preflight.py, harness_io.py, predictions.py, validation_budget.py, variant_promote.py
  • harness_default/: planner_catalog.yaml, planner_policy.yaml, manifest.yaml, gates.yaml
  • reports/: best-variant tables (csv/json/xlsx) and the summary scripts
  • skills/: int8-w8a8-gemm foundations/decode/prefill skill packs
  • tools/baseline/: Triton INT8 benchmark harness, baseline graph JSONs
  • 11 new test modules (planner, gate policy, gpu preflight, harness io, predictions, validation budget, variant promote, plan render, prompt schema, planner wiring, planner policy)
  • retuned variant kernels for glm52/TP8 and hy3/TP4, refreshed WORKFLOW.md

Changed:

  • gen_and_opt_pipeline.py, w8a8_pipeline.py, real_pipeline.py, pipeline.py, skill_store.py, variant_store.py, server/routes.py, static (detail view, shape input, css), form.yaml, bridge/dsh/dsh_agent.py, api/assets helpers
  • PR-side fixes kept: init_or_resume(task_id) alignment (73c213c) and the CPU-only-CI torch guards (3d70f58) / hermetic dsh tests (6a8f235)

Excluded: .bak- kernel backups.

Content imported from the DCU dev container tree (/root/zth_agent/MetaInfer),
3-way merged onto the merged PR HuangPuStar#26 head so the next-app alignment and CI fixes
stay in place. Scope is limited to this task plugin.

Added:
- orchestrator: planner.py (+ planner_policy/manifest/gates), gate_policy.py,
  gpu_preflight.py, harness_io.py, predictions.py, validation_budget.py,
  variant_promote.py
- harness_default/: planner_catalog.yaml, planner_policy.yaml, manifest.yaml,
  gates.yaml
- reports/: best-variant tables (csv/json/xlsx) and the summary scripts
- skills/: int8-w8a8-gemm foundations/decode/prefill skill packs
- tools/baseline/: Triton INT8 benchmark harness, baseline graph JSONs
- 11 new test modules (planner, gate policy, gpu preflight, harness io,
  predictions, validation budget, variant promote, plan render, prompt schema,
  planner wiring, planner policy)
- retuned variant kernels for glm52/TP8 and hy3/TP4, refreshed WORKFLOW.md

Changed:
- gen_and_opt_pipeline.py, w8a8_pipeline.py, real_pipeline.py, pipeline.py,
  skill_store.py, variant_store.py, server/routes.py, static (detail view,
  shape input, css), form.yaml, bridge/dsh/dsh_agent.py, api/assets helpers
- PR-side fixes kept: init_or_resume(task_id) alignment (73c213c) and the
  CPU-only-CI torch guards (3d70f58) / hermetic dsh tests (6a8f235)

Excluded: *.bak-* kernel backups.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant