| --- |
| license: apache-2.0 |
| task_categories: |
| - other |
| language: |
| - en |
| tags: |
| - triton |
| - gpu-kernels |
| - verification |
| - numerical-analysis |
| - reward-model |
| - rl-environments |
| pretty_name: Litmus Kernel Verification Corpus |
| size_categories: |
| - n<1K |
| configs: |
| - config_name: kernels |
| data_files: kernels.jsonl |
| - config_name: bug_pairs |
| data_files: bug_pairs.jsonl |
| --- |
| |
| # Litmus Kernel Verification Corpus |
|
|
| Correct and deliberately-broken Triton kernels, each broken one shipped **with |
| the input that exposes it**. |
|
|
| The corpus exists to measure one thing: how much of what a fixed-shape |
| `torch.rand()` `allclose` test calls "correct" actually is. On this corpus the |
| answer is that **88% of the planted bugs pass that test**. |
|
|
| ## Columns |
|
|
| | column | meaning | |
| |---|---| |
| | `name` | kernel identifier | |
| | `family` | elementwise / reduction / softmax / layernorm / matmul / attention | |
| | `bug_class` | `none` for correct kernels, otherwise one of six planted defect families | |
| | `triton_source` | the kernel, annotated with its buffer contract | |
| | `correct` | ground-truth label | |
| | `numerically_observable` | `false` when the defect has **no** numerical signature at any shape | |
| | `why_standard_misses` | why the fixed-shape `allclose` test lets this bug through | |
| | `witness_inputs` | JSON arrays: the exposing input (buggy kernels only) | |
| | `witness_dims` | the shape the witness uses | |
| | `witness_max_abs_error` | deviation from the float64 oracle on the witness | |
| | `standard_test_detects` | did `allclose(rtol=1e-2)` on the fixed shape catch it? | |
| | `litmus_verdict` | `accept` / `reject` / `unknown` at build time | |
|
|
| ## Bug families |
|
|
| - **`dtype_promotion`** — 2 kernels |
| - **`mask_leak`** — 4 kernels |
| - **`no_max_subtract`** — 2 kernels |
| - **`race`** — 3 kernels |
| - **`reduction_init`** — 3 kernels |
| - **`stride`** — 2 kernels |
| |
| ## Caveat |
| |
| The corpus is deliberately bug-heavy (16 buggy, 9 correct), so the |
| absolute "fraction correct" is a property of this corpus, not of kernels in |
| general. The finding is the **gap** between the standard test's pass rate and |
| the truth. |
| |
| ## Reproduce |
| |
| ```bash |
| git clone https://github.com/NagaYu/litmus && cd litmus |
| pip install -e ".[dev]" |
| python benchmarks/run.py # regenerates the numbers above |
| ``` |
| |
| Built by [Litmus](https://github.com/NagaYu/litmus) -- error bounds, symbolic GPU safety checks and targeted |
| falsification for Triton kernels. Interactive demo: [Space](https://huggingface.co/spaces/NagaYu/litmus). |
| |
