Hipfire Kernel Atlas

hipfire-kernel-atlas

Thin agent wrapper around scripts/kernel_atlas.py and .agents/skills/hipfire-kernel-atlas/render-fit.sh.

Canonical methodology (phases, corpus layout, ISA/dispatch manifests, task/eval loop): docs/methodology/kernel-atlas.md.

Promotion / correctness routes: docs/VALIDATION.md and, for timed claims, docs/methodology/perf-benchmarking.md. Atlas does not replace those owners.

What Atlas is

Treat every row as measured evidence tied to binary/prompt/git identity in the row. Dirty worktrees: cite provenance.diff_md5 and do not compare as a shipped baseline.

Producer / renderer surface

CLI (verify with python3 scripts/kernel_atlas.py --help):

Renderer wrapper (repo-root aware):

.agents/skills/hipfire-kernel-atlas/render-fit.sh \
  --row .codeinsight+research/kernel-atlas/runs/atlas.jsonl \
  --row-index 0 \
  --isa .codeinsight+research/kernel-atlas/runs/isa.json
# --dispatch optional; also optional if the row already references manifests

Raw corpus stays private/ignored:

mkdir -p .codeinsight+research/kernel-atlas/runs
mkdir -p .codeinsight+research/kernel-atlas/tasks

Workflow

1. Collect or locate rows under .codeinsight+research/kernel-atlas/runs/. Prefer existing JSONL before re-collecting.
2. Attach ISA with --isa-file or --isa-dir + --isa-filter; prefer --isa-output <path>.json so many rows share one manifest. Needs ROCm LLVM tools when inspecting HSACO (clang-offload-bundler, llvm-readobj, llvm-objdump).
3. Attach dispatch provenance with --dispatch-provenance and --dispatch-output when profiled kernel names exist. Evidence to inspect, not proof of one branch. Ranking is arch-aware when arch-specific sources exist (e.g. *.gfx1201.hip).
4. Render via render-fit.sh. With artifacts.profile_kernels, the view joins profiled names to ISA symbols and scopes the summary; unmatched hot names are printed on purpose.
5. suggest → experiment queue only. Auto-loads history from .codeinsight+research/kernel-atlas/tasks/ unless extra --history paths.
6. task → task.json + TASK.md. Pass --allowed-file for every editable path. Pass --correctness-command only as a claim-scoped command that still exists and matches docs/VALIDATION.md for that change (path-specific oracle, test_kernels, serve harness, etc.). Do not treat retired batteries as acceptance; use VALIDATION.
7. eval → refresh baseline first when the row carried profiling env (--refresh-baseline); compare candidates with --baseline. Status needs_baseline or unstable → no speedup claim. Ledger is local lineage, not a public benchmark.

Example commands

Paths and model files must exist on the machine; swap tags/files from registry/models.json.

Collect AR smoke with ISA + dispatch (illustrative; small model for speed):

python3 scripts/kernel_atlas.py collect-ar \
  --model ~/.hipfire/models/qwen3.5-0.8b.mq4 \
  --workload qwen3.5-0.8b \
  --model-size 0.8b \
  --quant mq4 \
  --prefill 32 \
  --gen 5 \
  --kv-mode asym3 \
  --profile-prefill \
  --profile-decode \
  --isa-dir .hipfire_kernels \
  --isa-filter 'gemm_hfq4g256|gemv_hfq4g256' \
  --isa-output .codeinsight+research/kernel-atlas/runs/isa.json \
  --dispatch-provenance \
  --dispatch-output .codeinsight+research/kernel-atlas/runs/dispatch.json \
  --output .codeinsight+research/kernel-atlas/runs/atlas.jsonl

DFlash collection (acceptance fixture = Qwen3.8-27B MQ4XT + measured draft; prompts under benchmarks/prompts/ when present):

python3 scripts/kernel_atlas.py collect-dflash \
  --target ~/.hipfire/models/qwen3.8-27b.mq4-xt \
  --draft ~/qcal/ladder-v2/drafts/qwen3.8-27b-dflash.mq4v2.hfq \
  --prompt-file benchmarks/prompts/merge_sort_thinking_off.txt \
  --workload qwen3.8-27b-mq4-xt-dflash-merge-sort \
  --max-tokens 256 \
  --ctx 2048 \
  --kv-mode q8 \
  --output .codeinsight+research/kernel-atlas/runs/atlas-dflash.jsonl

Suggest / task / eval:

python3 scripts/kernel_atlas.py suggest \
  --row .codeinsight+research/kernel-atlas/runs/atlas.jsonl \
  --row-index 1 \
  --isa .codeinsight+research/kernel-atlas/runs/isa.json \
  --dispatch .codeinsight+research/kernel-atlas/runs/dispatch.json \
  --format markdown

python3 scripts/kernel_atlas.py task \
  --row .codeinsight+research/kernel-atlas/runs/atlas.jsonl \
  --row-index 1 \
  --isa .codeinsight+research/kernel-atlas/runs/isa.json \
  --dispatch .codeinsight+research/kernel-atlas/runs/dispatch.json \
  --allowed-file kernels/src/gemv_hfq4g256_multirow.hip \
  --output-dir .codeinsight+research/kernel-atlas/tasks/example-gemv

python3 scripts/kernel_atlas.py eval \
  --task .codeinsight+research/kernel-atlas/tasks/example-gemv/task.json \
  --runs 5 --warmup-runs 1 \
  --refresh-baseline \
  --output-dir .codeinsight+research/kernel-atlas/tasks/example-gemv/eval-baseline

python3 scripts/kernel_atlas.py eval \
  --task .codeinsight+research/kernel-atlas/tasks/example-gemv/task.json \
  --baseline .codeinsight+research/kernel-atlas/tasks/example-gemv/eval-baseline/baseline.json \
  --runs 5 --warmup-runs 1 \
  --output-dir .codeinsight+research/kernel-atlas/tasks/example-gemv/eval-001

PyTorch-shape task shell (no automatic PyTorch kernel extract yet):

python3 scripts/kernel_atlas.py task-pytorch \
  --name example-rmsnorm-shape \
  --op rmsnorm \
  --input-shape 1,2048,4096 \
  --dtype float16 \
  --eval-command 'python3 bench_rmsnorm.py' \
  --allowed-file kernels/src/rmsnorm_candidate.hip \
  --output-dir .codeinsight+research/kernel-atlas/tasks/example-rmsnorm-shape

Interpretation rules

• ISA fit ≠ occupancy. Counters, residency, clocks, cache, and launch overlap are out of band unless separately measured.
• Matrix units present with zero observed matrix ops → ask whether the phase should use WMMA/MFMA or is a memory/launch-dominated decode GEMV.
• High VGPR/SGPR/spills → register pressure before bandwidth narratives.
• DFlash rows: tok/s is not correctness. Use VALIDATION’s claim-scoped route (serve semantics, path oracle, etc.) — not Atlas alone, and not retired batteries as current acceptance.
• eval unstable → measurement failure until run shape / thermal / DPM settles.
• Multi-host (e.g. hiptrx vs another box): compare only when prompt, binary md5, git diff md5, model, and variant env match; pin arch with HIPFIRE_TARGET_ARCH / ROCR_VISIBLE_DEVICES when multiple GPUs are visible.

After Atlas, before any ship claim

1. Rerun the VALIDATION route for the actual claim class (kernel channel, path oracle, serve harness, speed-gate / probe, Redline ladder — as applicable).
2. Follow perf protocol for any tok/s delta (fresh process, warmup, identity).
3. Do not write Atlas medians into product docs as floors or admissions.
4. Kernel-edit workflow after a hot kernel is identified: skill hipfire-kernel-tuning. New ISA targets: hipfire-arch-port.

Good agent output

Include:

• rendered fit section (or path to full render)
• row path, row index, ISA/dispatch manifest paths
• arch, quant, phase, shape bucket
• runtime metric used for the readout
• one short reading of likely limit / left on table
• explicit label: measured evidence, not promotion proof

Avoid:

• calling the heuristic a roofline model
• claiming a perf win from smoke or single dirty-tree rows
• mixing prompts/binaries without saying so
• citing Atlas success as merge/admission authority