[CI] pin ruff and make its configuration explicit

.github/scripts/compare_benchmark.py

@@ -36,7 +36,6 @@
 import csv
 import sys
 from pathlib import Path
-from typing import Dict, Tuple
 
 METRIC = "us"
 KERNEL_COLS = ("kernelName1", "kernelName2")
@@ -108,8 +107,8 @@ def _natural_key(val: str):
             return (1, val)
 
 
-Row = Dict[str, str]
-Key = Tuple[Tuple[str, str], ...]
+Row = dict[str, str]
+Key = tuple[tuple[str, str], ...]
 
 
 def _normalize_row(raw: Row) -> Row:
@@ -121,7 +120,7 @@ def _normalize_row(raw: Row) -> Row:
     }
 
 
-def _read_csv_rows(path: Path) -> Tuple[list[Row], Tuple[str, ...]]:
+def _read_csv_rows(path: Path) -> tuple[list[Row], tuple[str, ...]]:
     """Return (rows, fieldnames). Whitespace stripped from values."""
     if not path.exists():
         raise SystemExit(f"input csv not found: {path}")
@@ -136,7 +135,7 @@ def _read_csv_rows(path: Path) -> Tuple[list[Row], Tuple[str, ...]]:
     return rows, tuple(reader.fieldnames)
 
 
-def _key_cols(base_cols: Tuple[str, ...], cur_cols: Tuple[str, ...]) -> Tuple[str, ...]:
+def _key_cols(base_cols: tuple[str, ...], cur_cols: tuple[str, ...]) -> tuple[str, ...]:
     """Stable key column order across baseline/current schema drift."""
     cols = []
     seen = set()
@@ -148,8 +147,8 @@ def _key_cols(base_cols: Tuple[str, ...], cur_cols: Tuple[str, ...]) -> Tuple[st
     return tuple(cols)
 
 
-def _index_rows(rows: list[Row], key_cols: Tuple[str, ...]) -> Dict[Key, Row]:
-    indexed: Dict[Key, Row] = {}
+def _index_rows(rows: list[Row], key_cols: tuple[str, ...]) -> dict[Key, Row]:
+    indexed: dict[Key, Row] = {}
     for row in rows:
         key = tuple(sorted((c, row.get(c, "")) for c in key_cols))
         indexed[key] = row
@@ -167,7 +166,7 @@ def _parse_us(raw: Row) -> float | None:
 
 
 def _fmt_key_compact(
-    key: Key, key_cols_order: Tuple[str, ...], constants: Dict[str, str]
+    key: Key, key_cols_order: tuple[str, ...], constants: dict[str, str]
 ) -> str:
     """Format key showing only cols whose value is NOT in `constants`."""
     d = dict(key)
@@ -178,7 +177,7 @@ def _fmt_key_compact(
     return " ".join(parts) if parts else "(common)"
 
 
-def _find_constants(keys: list[Key], key_cols_order: Tuple[str, ...]) -> Dict[str, str]:
+def _find_constants(keys: list[Key], key_cols_order: tuple[str, ...]) -> dict[str, str]:
     """Return cols whose value is identical across all `keys`."""
     if not keys:
         return {}
@@ -304,7 +303,7 @@ def _entry_sort_key(e):
         if c not in HIDE_DISPLAY_COLS:
             display_cols.append(c)
 
-    def _cell_value(c: str, d: Dict[str, str]) -> str:
+    def _cell_value(c: str, d: dict[str, str]) -> str:
         if c in DERIVED_TUPLE_COLS:
             srcs = DERIVED_TUPLE_COLS[c]
             return "(" + ", ".join(d.get(s, "") for s in srcs) + ")"

.github/workflows/pre-checks.yaml

@@ -45,7 +45,11 @@ jobs:
         with:
           python-version: "3.12"
       - name: Install dependencies
-        run: pip3 install ruff
+        # Pinned on purpose. Unpinned, a ruff release that widens the default
+        # rule set turns unrelated PRs red with no change on our side (0.16.0
+        # did exactly that). Bump this deliberately, with the resulting
+        # findings burned down in the same or a follow-up PR.
+        run: pip3 install ruff==0.16.0
       - name: Install reviewdog
         uses: reviewdog/action-setup@e04ffabe3898a0af8d0fb1af00c188831c4b5893
       - name: Run ruff with reviewdog

aiter/__init__.py

@@ -1,10 +1,11 @@
 # SPDX-License-Identifier: MIT
 # Copyright (C) 2024-2026, Advanced Micro Devices, Inc. All rights reserved.
 
-import torch
+import logging
 import os
 import sys
-import logging
+
+import torch
 
 logger = logging.getLogger("aiter")
 
@@ -73,7 +74,7 @@ def getLogger():
 if AITER_TRITON_ONLY:
     logger.info("Triton ops only: CK and HIP ops (and their JIT build) are skipped.")
 elif AITER_AOT_IMPORT:
-    from .jit import core as core  # noqa: E402
+    from .jit import core as core
 else:
     # NOTE: do NOT wrap this block in try/except.
     # Catching ImportError here silently truncates the top-level aiter
@@ -84,62 +85,71 @@ def getLogger():
     # opus is gfx950-only but the package self-guards (warn + stubs on
     # non-gfx950) inside aiter/ops/opus/__init__.py, so its import line
     # is safe to put at top-level without try/except.
-    from .jit import core as core  # noqa: E402
-    from .utility import dtypes as dtypes  # noqa: E402
-    from .ops.enum import *  # noqa: F403,E402
-    from .ops.norm import *  # noqa: F403,E402
-    from .ops.quant import *  # noqa: F403,E402
-    from .ops.gemm_op_a8w8 import *  # noqa: F403,E402
-    from .ops.gemm_op_a16w16 import *  # noqa: F403,E402
-    from .ops.gemm_op_a4w4 import *  # noqa: F403,E402
-    from .ops.gemm_op_a8w4 import *  # noqa: F403,E402
-    from .ops.batched_gemm_op_a8w8 import *  # noqa: F403,E402
-    from .ops.batched_gemm_op_bf16 import *  # noqa: F403,E402
-    from .ops.deepgemm import *  # noqa: F403,E402
-    from .ops.opus import *  # noqa: F403,E402
-    from .ops.aiter_operator import *  # noqa: F403,E402
-    from .ops.activation import *  # noqa: F403,E402
-    from .ops.attention import *  # noqa: F403,E402
-    from .ops.custom import *  # noqa: F403,E402
-    from .ops.custom_all_reduce import *  # noqa: F403,E402
-    from .ops.quick_all_reduce import *  # noqa: F403,E402
-    from .ops.moe_op import *  # noqa: F403,E402
-    from .ops.moe_sorting import *  # noqa: F403,E402
-    from .ops.moe_sorting_opus import *  # noqa: F403,E402
-    from .ops.moe_mxfp4_aux import *  # noqa: F403,E402
-    from .ops.pa_sparse_prefill_opus import *  # noqa: F403,E402
-    from .ops.pos_encoding import *  # noqa: F403,E402
-    from .ops.cache import *  # noqa: F403,E402
-    from .ops.rmsnorm import *  # noqa: F403,E402
-    from .ops.communication import *  # noqa: F403,E402
-    from .ops.rope import *  # noqa: F403,E402
-    from .ops.topk import *  # noqa: F403,E402
-    from .ops.topk_plain import topk_plain  # noqa: F403,F401,E402
-    from .ops.mha import *  # noqa: F403,E402
-    from .ops.gradlib import *  # noqa: F403,E402
-    from .ops.trans_ragged_layout import *  # noqa: F403,E402
-    from .ops.sample import *  # noqa: F403,E402
-    from .ops.fused_qk_norm_mrope_cache_quant import *  # noqa: F403,E402
-    from .ops.fused_qknorm_idxrqknorm import (  # noqa: F401,E402
+    # isort: off
+    # Order below is load-bearing, do not sort. `dtypes` must be bound on the
+    # aiter package before any submodule that does `from aiter import dtypes`
+    # (mla.py among others) gets pulled in, otherwise that resolves against a
+    # partially initialised aiter and raises ImportError.
+    from .jit import core as core
+    from .utility import dtypes as dtypes
+    from .ops.enum import *
+    from .ops.norm import *
+    from .ops.quant import *
+    from .ops.gemm_op_a8w8 import *
+    from .ops.gemm_op_a16w16 import *
+    from .ops.gemm_op_a4w4 import *
+    from .ops.gemm_op_a8w4 import *
+    from .ops.batched_gemm_op_a8w8 import *
+    from .ops.batched_gemm_op_bf16 import *
+    from .ops.deepgemm import *
+    from .ops.opus import *
+    from .ops.aiter_operator import *
+    from .ops.activation import *
+    from .ops.attention import *
+    from .ops.custom import *
+    from .ops.custom_all_reduce import *
+    from .ops.quick_all_reduce import *
+    from .ops.moe_op import *
+    from .ops.moe_sorting import *
+    from .ops.moe_sorting_opus import *
+    from .ops.moe_mxfp4_aux import *
+    from .ops.pa_sparse_prefill_opus import *
+    from .ops.pos_encoding import *
+    from .ops.cache import *
+    from .ops.rmsnorm import *
+    from .ops.communication import *
+    from .ops.rope import *
+    from .ops.topk import *
+    from .ops.topk_plain import topk_plain  # noqa: F401
+    from .ops.mha import *
+    from .ops.gradlib import *
+    from .ops.trans_ragged_layout import *
+    from .ops.sample import *
+    from .ops.fused_qk_norm_mrope_cache_quant import *
+    from .ops.fused_qknorm_idxrqknorm import (  # noqa: F401
         fused_qknorm_idxrqknorm,
     )
-    from .ops.fused_qk_norm_rope_cache_quant import *  # noqa: F403,E402
-    from .ops.fused_qk_rmsnorm_group_quant import *  # noqa: F403,E402
-    from .ops.groupnorm import *  # noqa: F403,E402
-    from .ops.mhc import *  # noqa: F403,E402
-    from .ops.causal_conv1d_update import *  # noqa: F403,E402
-    from .ops.fused_split_gdr_update import *  # noqa: F403,E402
-    from . import mla  # noqa: F403,F401,E402
+    from .ops.fused_qk_norm_rope_cache_quant import *
+    from .ops.fused_qk_rmsnorm_group_quant import *
+    from .ops.groupnorm import *
+    from .ops.mhc import *
+    from .ops.causal_conv1d_update import *
+    from .ops.fused_split_gdr_update import *
+    from . import mla  # noqa: F401
+
+    # isort: on
 
 # Import Triton-based communication primitives from ops.triton.comms (optional, only if Iris is available)
 try:
     from .ops.triton.comms import (
+        IRIS_COMM_AVAILABLE,
         IrisCommContext,  # noqa: F401
-        calculate_heap_size,  # noqa: F401
-        reduce_scatter as iris_reduce_scatter,  # noqa: F401  # avoid shadowing C++ reduce_scatter exported by custom_all_reduce.py
         all_gather,  # noqa: F401
+        calculate_heap_size,  # noqa: F401
         reduce_scatter_rmsnorm_quant_all_gather,  # noqa: F401
-        IRIS_COMM_AVAILABLE,  # noqa: F401
+    )
+    from .ops.triton.comms import (
+        reduce_scatter as iris_reduce_scatter,  # noqa: F401  # avoid shadowing C++ reduce_scatter exported by custom_all_reduce.py
     )
 except (ImportError, AttributeError):
     # Iris or triton not available, skip import

aiter/aot/asm_mla_decode_fwd.py

@@ -1,6 +1,7 @@
-from collections import namedtuple
-import os
 import concurrent.futures
+import os
+from collections import namedtuple
+
 from csrc.cpp_itfs.mla.asm_mla_decode_fwd import compile
 from csrc.cpp_itfs.utils import AITER_CORE_DIR
 

aiter/aot/flydsl/common.py

@@ -5,17 +5,18 @@
 
 from __future__ import annotations
 
-from contextlib import contextmanager
+import enum
 import json
 import multiprocessing
-from multiprocessing.connection import wait as wait_for_sentinels
+import os
 import shutil
 import tempfile
 import time
+from collections.abc import Callable, Iterator
+from contextlib import contextmanager
 from dataclasses import dataclass
-import enum
-import os
-from typing import Any, Callable, Iterator
+from multiprocessing.connection import wait as wait_for_sentinels
+from typing import Any
 
 _DEFAULT_KERNEL_TIMEOUT = 1200.0
 _DEFAULT_MAX_WORKERS = 64

aiter/aot/flydsl/gemm.py

@@ -34,7 +34,6 @@
 import re
 import sys
 import time
-from typing import Dict, Optional
 
 import flydsl.expr as fx
 
@@ -83,7 +82,7 @@
 }
 
 
-def _parse_bool(value: Optional[str]) -> bool:
+def _parse_bool(value: str | None) -> bool:
     if value is None:
         return False
     normalized = value.strip().lower()
@@ -96,7 +95,7 @@ def _parse_bool(value: Optional[str]) -> bool:
     raise ValueError(f"Expected True/False, got {value!r}")
 
 
-def _parse_preshuffle_kernel_name(name: str) -> Optional[Dict]:
+def _parse_preshuffle_kernel_name(name: str) -> dict | None:
     m = _PRESHUFFLE_RE.fullmatch(name)
     if m is None:
         return None

aiter/aot/flydsl/grouped_moe.py

@@ -439,23 +439,23 @@ def compile_one_config(**job):
         )
         warp_tile_m = job["tile_m"] // job["m_warp"]
         contiguous = bool(job.get("grouped_contiguous_m", False))
-        common = dict(
-            model_dim=job["model_dim"],
-            inter_dim=job["inter_dim"],
-            experts=job["experts"],
-            max_m=job["max_m"],
-            tile_m=job["tile_m"],
-            tile_n=job["tile_n"],
-            tile_k=job["tile_k"],
-            m_warp=job["m_warp"],
-            n_warp=job["n_warp"],
-            out_dtype=job["out_dtype"],
-            num_buffers=job["num_buffers"],
-            grouped_persistent_m=job["grouped_persistent_m"],
-            grouped_contiguous_m=contiguous,
-            persistent_workers=job["persistent_workers"],
-            expert_sched_mode=job["expert_sched_mode"],
-        )
+        common = {
+            "model_dim": job["model_dim"],
+            "inter_dim": job["inter_dim"],
+            "experts": job["experts"],
+            "max_m": job["max_m"],
+            "tile_m": job["tile_m"],
+            "tile_n": job["tile_n"],
+            "tile_k": job["tile_k"],
+            "m_warp": job["m_warp"],
+            "n_warp": job["n_warp"],
+            "out_dtype": job["out_dtype"],
+            "num_buffers": job["num_buffers"],
+            "grouped_persistent_m": job["grouped_persistent_m"],
+            "grouped_contiguous_m": contiguous,
+            "persistent_workers": job["persistent_workers"],
+            "expert_sched_mode": job["expert_sched_mode"],
+        }
         if contiguous:
             act_lead = 1
             ub = job["token_num"] * job["topk"] + job["experts"] * (job["tile_m"] - 1)

aiter/aot/flydsl/moe.py

@@ -27,7 +27,6 @@
 import os
 import sys
 import time
-from typing import Optional
 
 from aiter.aot.flydsl.common import (
     collect_aot_jobs,
@@ -202,7 +201,7 @@ def _precompile_to_cache(
     # pin ``waves_per_eu`` in ``get_flydsl_stage{1,2}_kernels`` (only the
     # production-variant ``_persist_async_w4_cumul3`` does), causing
     # ``AOT cache miss`` at runtime even though the .pkl is present on disk.
-    waves_per_eu: Optional[int] = None,
+    waves_per_eu: int | None = None,
     k_batch: int = 1,
     b_nt: int = 2,
     gate_mode: str = "separated",

aiter/aot/flydsl/mxfp4_moe.py

@@ -57,12 +57,12 @@ def _job_key(job: dict) -> tuple:
 
 def parse_csv(csv_path: str):
     """Parse an fp4 tuned CSV into unique mxmoe-port compile jobs (one per stage)."""
+    from aiter.ops.flydsl.mxfp4_gemm2_kernels import _epilog_of
     from aiter.ops.flydsl.mxfp4_kname import (
         _is_mxfp4_kname,
         _parse_mxfp4_g1_kname,
         _parse_mxfp4_g2_kname,
     )
-    from aiter.ops.flydsl.mxfp4_gemm2_kernels import _epilog_of
 
     jobs = []
     seen = set()

aiter/aot/pa.py

@@ -1,6 +1,7 @@
-from collections import namedtuple
-import os
 import concurrent.futures
+import os
+from collections import namedtuple
+
 from csrc.cpp_itfs.pa.pa import compile
 
 PAConfig = namedtuple(

aiter/aot/pa_ragged.py

@@ -1,6 +1,7 @@
-from collections import namedtuple
-import os
 import concurrent.futures
+import os
+from collections import namedtuple
+
 from csrc.cpp_itfs.pa.pa_ragged import compile
 
 PAConfig = namedtuple(

aiter/aot/pa_v1.py

@@ -1,6 +1,7 @@
-from collections import namedtuple
-import os
 import concurrent.futures
+import os
+from collections import namedtuple
+
 from csrc.cpp_itfs.pa.pa_v1 import compile
 
 PAConfig = namedtuple(