CSV-canonical for the 3 target-map dashboards - #65
Conversation
Closes the data-shaped CSV-canonical tail (COMPILER_AUDIT). The per-target capability matrices were the last row-table dashboards still byte-gated on Markdown; they now emit a canonical machine-readable CSV like the other 9. - apple_target_map.render_csv() — one row per op (op_name, family, cpu/gpu status/framework/dtypes/symbol/dispatch, proof_test, notes), sorted by (family, op_name); dtype lists comma-joined in a quoted cell. - gpu_target_map.render_csv(target) — per-target rows (op_name, family, status, dtypes, arch_min, tile_shape, expected_mfu, roofline, notes). - generated_docs registry: the 3 target-map entries gain csv_path + render_csv, so the CSV becomes the drift-gated artifact and the Markdown the human companion (verified: a CSV edit now trips the drift gate). 12 dashboards are now CSV-canonical; the remaining markdown-only docs are narrative rollups, not row tables. Content consolidation stays deliberately deferred (Next Work #6 reassessment). Tests: test_generated_docs_registry + test_apple_target_map green (40); drift gate in sync; mypy clean. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 554fd29104
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| csv_path=_GEN / "verifier_coverage.csv", render_csv=_r_verifier_csv, | ||
| ), | ||
| # ── Target maps ── | ||
| # ── Target maps (CSV-canonical 2026-06-11) ── |
There was a problem hiding this comment.
When these target maps become CSV-canonical here, the existing per-target workflows still only operate on Markdown: tessera.cli.apple_target_map --render and tessera.cli.gpu_target_map --render call write_doc() which writes only .md, while scripts/validate.sh still runs the GPU CLI --check against .md. In the scenario where a target-map source changes and a developer follows the old CLI/test failure message, the Markdown gets refreshed but the new canonical CSV remains stale, so python -m tessera.compiler.generated_docs --check still fails. Please either route those CLIs/checks through generated_docs or teach them to write/check the CSV artifact too.
Useful? React with 👍 / 👎.
1 participant
Closes the data-shaped CSV-canonical tail (COMPILER_AUDIT). The per-target capability matrices (apple / nvidia_sm90 / rocm) were the last row-table dashboards still byte-gated on Markdown; they now emit a canonical machine-readable CSV like the other 9.
apple_target_map.render_csv()+gpu_target_map.render_csv(target)emit one row per op (stable sort by(family, op_name); dtype lists comma-joined in a quoted cell).csv_path+render_csv, so the CSV is the drift-gated artifact and the Markdown the human companion — verified a CSV edit now trips the drift gate.12 dashboards are now CSV-canonical; the remaining markdown-only docs are narrative rollups, not row tables. Content consolidation stays deliberately deferred (Next Work #6 reassessment).
Tests:
test_generated_docs_registry+test_apple_target_mapgreen (40); drift gate in sync; mypy clean.🤖 Generated with Claude Code