11.1

Adjusted Signal

11.1

Raw Pattern Score

100%

Time Factor

2026-10-31

Last Push

20.5K

Stars

MLIR

Language

442.5K

Lines of Code

1.3K

Files

4.5K

Pattern Hits

2026-10-31

Scan Date

0.03

HC Hit Rate

Risk Review · Payload v1

Use each vector to guide manual verification of the underlying findings.

Payload schema: 1.0.0 · Engine: 2.0.0 · Scoring: heuristic-density-v1 · Scan date: 2026-10-31.

Cognitive Ownership Risk 28.2/100 · MEDIUM

Can maintainers explain intent, assumptions, and trade-offs? 3287 finding(s).

Hallucinated Dependency & Supply Chain Risk 8.4/100 · LOW

Are imports declared and package/API names verified? 663 finding(s).

Agent Over-Engineering Ratio 26.5/100 · MEDIUM

Are wrappers, guards, and repeated structures necessary? 2493 finding(s).

PR Review Velocity Impact 47.1/100 · MEDIUM

Which changes need focused human review? 4466 finding(s).

What These Metrics Mean

Adjusted Signal: Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score: Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor: A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits: Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate: High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files: Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.

Severity Breakdown

Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.

CRITICAL 27HIGH 18MEDIUM 420LOW 4001

Directory Score Breakdown

This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.

Pattern Findings

The scanner identified 4466 distinct pattern matches across 25 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Hyper-Verbose Identifiers1939 hits · 1505 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	577	def get_triton_version_suffix():	CODE
LOW	python/conftest.py	60	def fresh_knobs_including_libraries():	CODE
LOW	python/build_helpers.py	76	def find_therock_rocm_include_dir() -> Optional[str]:	CODE
LOW	python/build_helpers.py	161	def _download_file_with_urllib(url: str, path: str, label: str):	CODE
LOW	python/build_helpers.py	454	def get_amd_codegen_package_info(helper_args: BuildHelperArgs):	CODE
LOW	python/build_helpers.py	490	def _get_thirdparty_package_cmake_vars(package: Package, helper_args: BuildHelperArgs):	CODE
LOW	python/build_helpers.py	524	def get_thirdparty_cmake_vars(packages: list[str], helper_args: BuildHelperArgs):	CODE
LOW	python/build_helpers.py	544	def write_thirdparty_cmake_vars(output: str, packages: list[str], helper_args: BuildHelperArgs):	CODE
LOW	python/build_helpers.py	641	def download_and_copy_amd_codegen(helper_args: BuildHelperArgs):	CODE
LOW	python/build_helpers.py	729	def get_nvidia_toolchain_packages():	CODE
LOW	python/build_helpers.py	803	def download_and_copy_dependencies(helper_args: BuildHelperArgs):	CODE
LOW	python/test/unit/test_utils.py	12	def test_cdiv_symbolic_numerator(y):	CODE
LOW	python/test/unit/test_utils.py	54	def test_validate_block_shape_rejects_zero():	CODE
LOW	python/test/unit/test_utils.py	63	def test_validate_block_shape_accepts_powers_of_two():	CODE
LOW	python/test/unit/test_utils.py	80	def test_get_dram_gbps_memory_data_rate(monkeypatch, backend, arch, properties, expected):	CODE
LOW	python/test/unit/test_utils.py	88	def test_perf_report_creates_save_path(tmp_path):	CODE
LOW	python/test/unit/test_debug.py	27	def _run_device_assert_barrier(device):	CODE
LOW	python/test/unit/test_debug.py	41	def _run_expect_zero_device_assert(device):	CODE
LOW	python/test/unit/test_debug.py	58	def test_expect_zero_device_assert(device):	CODE
LOW	python/test/unit/test_debug.py	84	def test_device_assert_barrier(device):	CODE
LOW	python/test/unit/test_debug.py	163	def test_sanitize_int_add_overflow(x, y, x_dtype, y_dtype, debug, should_overflow, device):	CODE
LOW	python/test/unit/test_debug.py	179	def test_sanitize_int_mul_overflow(x, y, x_dtype, y_dtype, debug, should_overflow, device):	CODE
LOW	python/test/unit/test_debug.py	194	def test_sanitize_int_sub_overflow(x, y, x_dtype, y_dtype, debug, should_overflow, device):	CODE
LOW	python/test/unit/test_perf_warning.py	11	def enable_diagnostics_context(value):	CODE
LOW	python/test/unit/test_perf_warning.py	95	def test_remark_vectorization(capfd, fresh_triton_cache):	CODE
LOW	python/test/unit/test_perf_warning.py	159	def test_remark_swp_op_before_operands(capfd, fresh_triton_cache):	CODE
LOW	python/test/unit/test_stages_inspection.py	11	def test_instrumentation_mode():	CODE
LOW	python/test/unit/test_stages_inspection.py	18	def test_dynamic_instrumentation_and_dialect_loading():	CODE
LOW	python/test/unit/test_knobs.py	73	def _register_pressure_scheduler_kernel():	CODE
LOW	python/test/unit/test_knobs.py	102	def test_nvidia_register_pressure_scheduler_hook(link_hip_first, fresh_triton_cache):	CODE
LOW	python/test/unit/test_knobs.py	116	def test_nvidia_register_pressure_scheduler_concurrent():	CODE
LOW	python/test/unit/test_knobs.py	133	def _nvidia_short_pointer_kernel():	CODE
LOW	python/test/unit/test_knobs.py	154	def test_nvidia_short_pointer_abi(link_hip_first, fresh_triton_cache):	STRING
LOW	python/test/unit/test_knobs.py	177	def _compile_empty_hip_kernel(arch, warp_size):	STRING
LOW	python/test/unit/test_knobs.py	207	def test_amd_llvm_options_concurrent():	STRING
LOW	python/test/unit/test_knobs.py	282	def test_amd_bitcode_path_override(fresh_knobs, monkeypatch, tmp_path):	STRING
LOW	python/test/unit/test_knobs.py	299	def test_amd_codegen_path_override(fresh_knobs, monkeypatch):	STRING
LOW	python/test/unit/test_knobs.py	311	def test_amd_codegen_revision_invalidates_backend_hash(fresh_knobs, monkeypatch):	STRING
LOW	python/test/unit/test_knobs.py	326	def test_amd_codegen_inlines_functions_from_bitcode(fresh_knobs, monkeypatch):	STRING
LOW	python/test/unit/test_knobs.py	366	def test_amd_codegen_options_restored(option, capfd, fresh_knobs):	STRING
LOW	python/test/unit/test_knobs.py	381	def test_amd_codegen_reports_invalid_llvm_ir(fresh_knobs):	STRING
LOW	python/test/unit/test_knobs.py	391	def test_amd_codegen_assembles_object_with_private_llvm(fresh_knobs):	STRING
LOW	python/test/unit/test_knobs.py	410	def test_amd_codegen_respects_mir_dump_knob(enabled, fresh_knobs, monkeypatch, tmp_path):	STRING
LOW	python/test/unit/test_knobs.py	433	def test_amd_codegen_preserves_mir_replacement(fresh_knobs, monkeypatch):	STRING
LOW	python/test/unit/test_knobs.py	458	def _loop_strength_reduction_kernel():	STRING
LOW	python/test/unit/test_knobs.py	502	def test_nvidia_lsr_concurrent():	STRING
LOW	python/test/unit/tools/test_aot.py	329	def compile_aot_kernel_no_specialization(dir, kernel_path, dtype, BM, BN, BK, target=None):	STRING
LOW	python/test/unit/tools/test_aot.py	374	def generate_matmul_test_data(dir, M, N, K):	STRING
LOW	python/test/unit/tools/test_aot.py	399	def test_linker_rejects_invalid_algo_id():	STRING
LOW	python/test/unit/tools/test_aot.py	465	def test_compile_link_matmul_no_specialization():	STRING
LOW	python/test/unit/tools/test_aot.py	532	def test_launcher_has_no_available_kernel():	STRING
LOW	python/test/unit/tools/test_aot.py	570	def test_compile_link_autotune_matmul():	STRING
LOW	python/test/unit/tools/test_aot.py	655	def test_aot_target_parsing_with_explicit_target():	STRING
LOW	python/test/unit/tools/test_aot.py	700	def gen_scalar_float_test_bin(dir, block, alpha, exe="test_scalar_float"):	STRING
LOW	python/test/unit/tools/test_aot.py	757	def test_compile_link_scalar_float(dtype, alpha):	STRING
LOW	python/test/unit/tools/test_linear_layout.py	36	def test_operator_mul_identity():	CODE
LOW	python/test/unit/tools/test_linear_layout.py	42	def test_operator_mul_disjoint_dims():	CODE
LOW⚡	python/test/unit/tools/test_linear_layout.py	75	def test_get_matrix_view_identity():	CODE
LOW⚡	python/test/unit/tools/test_linear_layout.py	83	def test_get_matrix_view_strided():	CODE
LOW⚡	python/test/unit/tools/test_linear_layout.py	92	def test_get_matrix_view_from_bases():	CODE
1879 more matches not shown…

Decorative Section Separators378 hits · 994 pts

Severity	File	Line	Snippet	Context
MEDIUM	python/test/unit/cuda/test_libdevice_cuda.py	13	# -----------------------	COMMENT
MEDIUM	python/test/unit/cuda/test_libdevice_cuda.py	15	# -----------------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	12	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	14	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	84	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	86	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	152	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	154	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	252	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_standard.py	254	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	376	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	378	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	608	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	674	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	676	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	702	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	704	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	734	# ----------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	736	# ----------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	783	# ------------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	785	# ------------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	827	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	829	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	977	# ----------------------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	979	# ----------------------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	993	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	995	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1087	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1089	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1121	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1123	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1446	# -------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1448	# -------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1537	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1539	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1584	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1586	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1618	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1620	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1645	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1647	# ----------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1828	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	1830	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	2646	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	2648	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	3254	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	3256	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	3964	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	3966	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	4033	# ------------------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	4035	# ------------------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	4271	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	4273	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	4388	# ---------------	COMMENT
MEDIUM	python/test/unit/language/test_core.py	4390	# ---------------	COMMENT
MEDIUM⚡	python/test/unit/language/test_core.py	5904	# ---------------	STRING
MEDIUM⚡	python/test/unit/language/test_core.py	5906	# ---------------	STRING
MEDIUM⚡	python/test/unit/language/test_core.py	6121	# --------------------	STRING
MEDIUM⚡	python/test/unit/language/test_core.py	6123	# --------------------	STRING
MEDIUM⚡	python/test/unit/language/test_core.py	6361	# -------------	STRING
318 more matches not shown…

Over-Commented Block951 hits · 914 pts

Severity	File	Line	Snippet	Context
LOW	unittest/Tools/LLVMOptionsTest.cpp	1	#include "triton/Tools/LLVMOptions.h"	COMMENT
LOW	unittest/Analysis/UtilityTest.cpp	1	#include "triton/Dialect/Triton/IR/Utility.h"	COMMENT
LOW	unittest/Dialect/TritonGPU/SwizzleTest.cpp	1	#include "triton/Dialect/Triton/IR/Dialect.h"	COMMENT
LOW	unittest/Dialect/TritonGPU/SwizzleTest.cpp	241	int phaseIdx;	COMMENT
LOW	unittest/Dialect/TritonGPU/SwizzleTest.cpp	261	} else {	COMMENT
LOW	…test/Dialect/TritonGPU/LinearLayoutConversionsTest.cpp	1	#include "triton/Dialect/TritonGPU/IR/LinearLayoutConversions.h"	COMMENT
LOW	…test/Dialect/TritonGPU/LinearLayoutConversionsTest.cpp	3581	//	COMMENT
LOW	test/lib/Analysis/TestMembar.cpp	1	#include "../third_party/nvidia/include/TritonNVIDIAGPUToLLVM/Utility.h"	COMMENT
LOW	bin/triton-llvm-opt.cpp	1	/// Trimmed down clone of llvm opt to be able to test triton custom llvm ir	COMMENT
LOW	bin/triton-tensor-layout.cpp	1	#include "RegisterTritonDialects.h"	COMMENT
LOW	bin/triton-tensor-layout.cpp	21	//	COMMENT
LOW	bin/RegisterTritonDialects.h	1	#pragma once	COMMENT
LOW	bin/RegisterTritonDialects.h	21	// Below headers will allow registration to ROCm passes	COMMENT
LOW	python/test/unit/language/test_line_info.py	301	# CHECK: %offsets_1 = tt.splat %offset_0 : i32 -> tensor<16xi32> loc(#loc15)	COMMENT
LOW	python/test/unit/language/test_line_info.py	521	# CHECK: #loc = loc("{{.*}}":@DEF_LINE@:5)	COMMENT
LOW	python/test/unit/language/test_line_info.py	541	# CHECK: } loc(#loc)	COMMENT
LOW	python/test/gluon/test_frontend.py	881	_TENSOR_TUPLE_288 = TensorTuple((288, ), _TENSOR_TUPLE_LAYOUT, minimum_padding_size=64)	COMMENT
LOW	python/test/gluon/test_frontend.py	901		COMMENT
LOW	python/test/gluon/test_frontend.py	941	# CHECK: [[VALUE:%.*]] = tt.make_range {end = 128 : i32, start = 0 : i32} : tensor<128xi32, [[LAYOUT]]>	COMMENT
LOW	python/test/gluon/test_frontend.py	1081	# CHECK-NEXT: [[A:%.*]] = tt.make_range {end = 1 : i32, start = 0 : i32}	COMMENT
LOW	python/test/gluon/test_frontend.py	1141	@filecheck_test	COMMENT
LOW	python/test/backend/test_mir_stage.py	61	# so we key instructions by (opcode, ordinal-within-function) instead -- a	COMMENT
LOW	python/test/backend/test_mir_stage.py	321	# filtering. That is only correct because the MIR is dumped at the	COMMENT
LOW	python/triton/knobs.py	561	roctx_lib_path: env_opt_str = env_opt_str("TRITON_ROCTX_LIB_PATH")	COMMENT
LOW	python/triton/_utils.py	161	# iterable. Both have type "tuple" making it difficult to distinguish	COMMENT
LOW	python/triton/experimental/gsan/src/Compat.h	1	#pragma once	COMMENT
LOW	python/triton/experimental/gsan/src/AllocatorCommon.h	1	#pragma once	COMMENT
LOW	python/triton/experimental/gsan/src/AllocatorCommon.h	41	if (handleType == CU_MEM_HANDLE_TYPE_FABRIC)	COMMENT
LOW	…iton/experimental/gluon/language/amd/cdna5/_layouts.py	181	assert block_m % slice_m == 0 and block_n % slice_n == 0, \	COMMENT
LOW	…iton/experimental/gluon/language/amd/cdna5/_layouts.py	201	# M=2: w0 w1 <- first tile computed by w1 (reg=0)	COMMENT
LOW	…iton/experimental/gluon/language/amd/cdna5/_layouts.py	221	# efficient.	COMMENT
LOW	python/triton/tools/disasm.py	1	# MIT License	COMMENT
LOW	…triton/tools/triton_to_gluon_translator/ordered_set.py	1	"""	COMMENT
LOW	python/examples/gluon/06-overlapping-accumulator.py	61	# With 512 TMEM columns, double buffered BLOCK_N=256 accumulators fully consume TMEM	COMMENT
LOW	python/examples/gluon/06-overlapping-accumulator.py	81	# 3) New ping-pong style barrier guards the overlapped columns.	COMMENT
LOW	python/examples/gluon/06-overlapping-accumulator.py	101	#	COMMENT
LOW	python/examples/gluon/02-conv-dgrad.py	61	#	COMMENT
LOW	python/examples/gluon/02-conv-fprop.py	61		COMMENT
LOW	python/examples/gluon/02-conv-fprop.py	481	output = torch.empty((N, out_h, out_w, Co), device=input_tensor.device, dtype=TORCH_GEMM_DTYPE)	COMMENT
LOW	python/examples/gluon/02-conv-wgrad.py	61	# grad_W[Co, RSCi] = grad_out[M, Co]^T @ im2col(input)[M, RSCi]	COMMENT
LOW	python/triton_kernels/tests/test_reduce.py	221		COMMENT
LOW	python/triton_kernels/triton_kernels/matmul.py	61		COMMENT
LOW	python/triton_kernels/triton_kernels/matmul.py	221	assert out_ndim >= storage.data.ndim	COMMENT
LOW	…_kernels/tensor_details/layout_details/hopper_value.py	201	sizes += [a, b, c, s, d]	COMMENT
LOW	…riton_kernels/tensor_details/layout_details/strided.py	1	from dataclasses import dataclass	COMMENT
LOW	…n_kernels/tensor_details/layout_details/torch_utils.py	1	import torch	COMMENT
LOW	…n_kernels/tensor_details/layout_details/torch_utils.py	21	# if not is_fp4:	COMMENT
LOW	python/tutorials/02-fused-softmax.py	61		COMMENT
LOW	python/tutorials/02-fused-softmax.py	181		COMMENT
LOW	python/tutorials/08-grouped-gemm.py	1	"""	COMMENT
LOW	python/tutorials/04-low-memory-dropout.py	21	# The dropout operator was first introduced in [SRIVASTAVA2014]_ as a way to improve the performance	COMMENT
LOW	python/tutorials/04-low-memory-dropout.py	81	output = dropout(x, x_keep=x_keep, p=p)	COMMENT
LOW	python/tutorials/04-low-memory-dropout.py	161		COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	21	# -----------	COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	41	# for k in range(0, K, BLOCK_SIZE_K):	COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	61	# For a row-major 2D tensor :code:`X`, the memory location of :code:`X[i, j]` is given	COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	81	# offs_k = tl.arange(0, BLOCK_SIZE_K)	COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	101	#	COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	121	# # Number of programs ids along the N axis	COMMENT
LOW	python/tutorials/03-matrix-multiplication.py	281	# We will advance this pointer as we move in the K direction	COMMENT
891 more matches not shown…

Unused Imports622 hits · 496 pts

Severity	File	Line	Context
LOW	python/test/unit/runtime/test_build.py	1	CODE
LOW	python/test/unit/language/test_annotations.py	1	CODE
LOW	python/test/gsan/test_write_once.py	1	CODE
LOW	python/test/gsan/test_allocator.py	1	CODE
LOW	python/test/gsan/test_gsan_failures.py	1	CODE
LOW	python/test/gsan/test_symmetric_memory.py	1	CODE
LOW	python/test/gsan/test_gsan.py	1	CODE
LOW	python/test/gsan/test_write_once_failures.py	1	CODE
LOW	python/triton/_instrumentation.py	1	CODE
LOW	python/triton/knobs.py	1	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	8	CODE
LOW	python/triton/__init__.py	20	CODE
LOW	python/triton/__init__.py	21	CODE
LOW	python/triton/__init__.py	21	CODE
LOW	python/triton/__init__.py	22	CODE
LOW	python/triton/__init__.py	22	CODE
LOW	python/triton/__init__.py	23	CODE
LOW	python/triton/__init__.py	24	CODE
LOW	python/triton/__init__.py	27	CODE
LOW	python/triton/__init__.py	28	CODE
LOW	python/triton/_utils.py	1	CODE
LOW	python/triton/experimental/gsan/_profile_allocator.py	2	CODE
LOW	python/triton/experimental/gsan/profile.py	7	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/profile.py	21	CODE
LOW	python/triton/experimental/gsan/graph.py	7	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/__init__.py	1	CODE
LOW	python/triton/experimental/gsan/_testing_utils.py	1	CODE
LOW	python/triton/experimental/gsan/_allocator.py	1	CODE
LOW	python/triton/experimental/gsan/symmetric_memory.py	1	CODE
LOW	python/triton/experimental/gsan/_testing.py	1	CODE
LOW	python/triton/experimental/gsan/_stream_sync.py	1	CODE
LOW	python/triton/experimental/gsan/_utils.py	1	CODE
LOW	python/triton/experimental/gluon/__init__.py	1	CODE
LOW	python/triton/experimental/gluon/__init__.py	1	CODE
LOW	python/triton/experimental/gluon/__init__.py	1	CODE
LOW	python/triton/experimental/gluon/__init__.py	1	CODE
LOW	python/triton/experimental/gluon/__init__.py	2	CODE
562 more matches not shown…

Hallucination Indicators27 hits · 260 pts

Severity	File	Line	Snippet	Context
CRITICAL	python/test/unit/plugins/test_plugin.py	36	triton._C.libtriton.passes.plugin.extend_with(LIB)	CODE
CRITICAL	python/test/unit/plugins/test_custom_ops.py	67	triton._C.libtriton.ir.builder.extend_with(LIB)	CODE
CRITICAL	python/test/unit/plugins/override_helper.py	15	triton._C.libtriton.passes.plugin.extend_with(LIB)	CODE
CRITICAL	python/test/unit/language/test_core.py	5282	if not is_interpreter() and triton.runtime.driver.active.utils.get_device_properties(	CODE
CRITICAL	python/test/unit/language/test_matmul.py	143	shared_mem_avail = triton.runtime.driver.active.utils.get_device_properties(0)["max_shared_mem"]	CODE
CRITICAL	python/test/gsan/test_gsan.py	1963	properties = triton.runtime.driver.active.utils.get_device_properties(torch.cuda.current_device())	CODE
CRITICAL	python/test/gluon/test_frontend.py	1730	%0 = tt.call @triton.experimental.gluon.language.nvidia.ampere.mbarrier.allocate_mbarrier__cNone_cTrue() : () -> !tt	CODE
CRITICAL	python/test/gluon/test_frontend.py	1735	tt.func private @triton.experimental.gluon.language.nvidia.ampere.mbarrier.allocate_mbarrier__cNone_cTrue() -> !ttg.me	CODE
CRITICAL	python/test/gluon/test_frontend.py	4794	%13 = tt.call @triton.experimental.gluon.language._standard.zeros__Tc1T_cfp16_cAL() : () -> tensor<1xf16, #gluon.aut	CODE
CRITICAL	python/test/gluon/test_frontend.py	4822	tt.func private @triton.experimental.gluon.language._standard.zeros__Tc1T_cfp16_cAL() -> tensor<1xf16, #gluon.auto_enc	CODE
CRITICAL	python/test/gluon/test_frontend.py	4853	%1 = tt.call @triton.experimental.gluon.language._standard.zeros__Tc1T_cbf16_cAL() : () -> tensor<1xbf16, #gluon.aut	CODE
CRITICAL	python/test/gluon/test_frontend.py	4871	tt.func private @triton.experimental.gluon.language._standard.zeros__Tc1T_cbf16_cAL() -> tensor<1xbf16, #gluon.auto_en	CODE
CRITICAL	python/triton/_internal_testing.py	88	return triton.runtime.driver.active.utils.cuOccupancyMaxActiveClusters(kernel.function, kernel.metadata.shared,	CODE
CRITICAL	python/triton/language/core.py	3428	triton.runtime.driver.active.utils.set_printf_fifo_size(size_bytes)	STRING
CRITICAL	python/triton_kernels/triton_kernels/target_info.py	80	return triton.runtime.driver.active.utils.get_device_properties(0)["multiprocessor_count"]	CODE
CRITICAL	third_party/proton/proton/hooks/instrumentation.py	135	max_shared_mem = triton.runtime.driver.active.utils.get_device_properties(device)["max_shared_mem"]	CODE
CRITICAL	third_party/proton/proton/hooks/instrumentation.py	160	arch = triton.runtime.driver.active.utils.get_device_properties(device)["arch"].split(":")[0]	CODE
CRITICAL⚡	third_party/amd/python/test/test_amd_codegen.py	114	declare void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(15), i32)	CODE
CRITICAL⚡	third_party/amd/python/test/test_amd_codegen.py	118	call void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(15) @bar, i32 4)	CODE
CRITICAL	third_party/amd/python/test/test_amd_codegen.py	139	declare void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(15), i32)	CODE
CRITICAL	third_party/amd/python/test/test_amd_codegen.py	140	declare i32 @llvm.amdgcn.s.get.named.barrier.state(ptr addrspace(15))	CODE
CRITICAL	third_party/amd/python/test/test_amd_codegen.py	166	call void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(15) %bar, i32 4)	STRING
CRITICAL⚡	third_party/amd/python/test/test_amd_codegen.py	190	call void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(15) %bar, i32 4)	STRING
CRITICAL⚡	third_party/amd/python/test/test_amd_codegen.py	192	%state = call i32 @llvm.amdgcn.s.get.named.barrier.state(ptr addrspace(15) %bar)	STRING
CRITICAL	third_party/amd/python/test/test_amd_codegen.py	222	call void @llvm.amdgcn.s.barrier.signal.var(ptr addrspace(15) %frozen, i32 4)	STRING
CRITICAL	third_party/nvidia/backend/driver.py	231	cu_tensor_map = triton.runtime.driver.active.utils.fill_tma_descriptor_im2col(	CODE
CRITICAL	third_party/nvidia/backend/driver.py	246	cu_tensor_map = triton.runtime.driver.active.utils.fill_tma_descriptor_tiled(	CODE

Deep Nesting203 hits · 162 pts

Severity	File	Line	Context
LOW	setup.py	129	CODE
LOW	python/build_helpers.py	393	CODE
LOW	python/test/unit/tools/test_triton_to_gluon.py	564	CODE
LOW	python/test/unit/cuda/test_libdevice_cuda.py	135	CODE
LOW	python/test/unit/runtime/test_specialize.py	43	CODE
LOW	python/test/unit/language/test_conversions.py	16	CODE
LOW	python/test/unit/language/test_conversions.py	278	CODE
LOW	python/test/unit/language/test_frontend.py	282	CODE
LOW	python/test/unit/language/test_frontend.py	331	CODE
LOW	python/test/unit/language/test_frontend.py	2061	CODE
LOW	python/test/unit/language/test_frontend.py	285	CODE
LOW	python/test/unit/language/test_frontend.py	334	CODE
LOW	python/test/unit/language/test_core.py	394	CODE
LOW	python/test/unit/language/test_core.py	3610	CODE
LOW	python/test/unit/language/test_core.py	3792	CODE
LOW	python/test/unit/language/test_core.py	4393	CODE
LOW	python/test/unit/language/test_core.py	4574	CODE
LOW	python/test/unit/language/test_core.py	5515	CODE
LOW	python/test/unit/language/test_core.py	5740	CODE
LOW	python/test/unit/language/test_core.py	5911	CODE
LOW	python/test/unit/language/test_core.py	6371	CODE
LOW	python/test/unit/language/test_core.py	6888	CODE
LOW	python/test/unit/language/test_core.py	8353	CODE
LOW	python/test/unit/language/test_core.py	3614	CODE
LOW	python/test/unit/language/test_core.py	6374	CODE
LOW	python/test/unit/language/test_core.py	6891	CODE
LOW	python/test/unit/language/test_core.py	8356	CODE
LOW	python/test/unit/language/test_pipeliner.py	223	CODE
LOW	python/test/unit/language/test_subprocess.py	57	CODE
LOW	python/test/unit/language/test_matmul.py	258	CODE
LOW	python/test/unit/language/test_matmul.py	633	CODE
LOW	python/test/unit/language/test_matmul.py	772	CODE
LOW	python/test/unit/language/test_matmul.py	262	CODE
LOW	python/test/unit/language/test_matmul.py	821	CODE
LOW	python/test/unit/language/test_line_info.py	164	CODE
LOW	python/test/unit/language/test_line_info.py	212	CODE
LOW	python/test/unit/language/test_annotations.py	62	CODE
LOW	python/test/unit/language/print_helper.py	107	CODE
LOW	python/test/unit/language/test_tensor_descriptor.py	1565	CODE
LOW	python/test/unit/language/test_tensor_descriptor.py	1574	CODE
LOW	python/test/unit/backend/test_gsan_profile.py	333	CODE
LOW	python/test/unit/backend/test_gsan_profile.py	623	CODE
LOW	python/test/unit/backend/test_gsan_profile.py	1031	CODE
LOW	python/test/unit/backend/test_gsan_profile.py	1085	CODE
LOW	python/test/unit/backend/test_gsan_profile.py	1139	CODE
LOW	python/test/gsan/test_gsan_failures.py	172	CODE
LOW	python/test/gsan/test_gsan_failures.py	779	CODE
LOW	python/test/gsan/test_symmetric_memory.py	432	CODE
LOW	python/test/gsan/test_symmetric_memory.py	522	CODE
LOW	python/test/gsan/test_gsan.py	1616	CODE
LOW	python/test/gsan/test_write_once_failures.py	31	CODE
LOW	python/test/gsan/test_write_once_failures.py	70	CODE
LOW	python/test/gluon/test_lowerings.py	26	CODE
LOW	python/test/gluon/test_lowerings.py	983	CODE
LOW	python/test/gluon/test_lowerings.py	1106	CODE
LOW	python/test/gluon/test_lowerings.py	1127	CODE
LOW	python/test/gluon/test_frontend.py	1320	CODE
LOW	python/test/gluon/test_frontend.py	5615	CODE
LOW	python/test/gluon/test_frontend.py	5831	CODE
LOW	python/test/gluon/test_frontend.py	5618	CODE
143 more matches not shown…

Structural Annotation Overuse43 hits · 85 pts

Severity	File	Line	Snippet	Context
LOW⚡	README.md	266	# Step 1: Run the kernel once to dump kernel's IRs and ptx/amdgcn in $TRITON_DUMP_DIR	COMMENT
LOW⚡	README.md	267	# Step 2: Copy $TRITON_DUMP_DIR/<kernel_hash> to $TRITON_OVERRIDE_DIR	COMMENT
LOW⚡	README.md	268	# Step 3: Delete the stages that you do not want to override and modify the stage you do want to override	COMMENT
LOW⚡	README.md	269	# Step 4: Run the kernel again to see the overridden result	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	149	# Step 1: Extract patches for each output position	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	156	# Step 2: Stack patches into im2col matrix A (M=4, K=4)	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	167	# Step 3: Reshape filter into weight matrix W (Co=1, K=4)	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	171	# Step 4: Output = A @ W^T	COMMENT
LOW	…ude/triton/Conversion/TritonGPUToLLVM/TargetInfoBase.h	120	// NOTE: This function is used for backend debug. DO NOT DELETE.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	526	// Step 1: Right shift for 8 lanes.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	529	// Step 2: Right shift for 4 lanes.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	532	// Step 3: Right shift for 2 lanes.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	535	// Step 4: Right shift for 1 lane.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	538	// Step 5: Broadcast lane 15 of each row to all the lanes of its next row.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	541	// Step 6: Broadcast lane 31 to lane 32-63.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	546	// Step 7: Read the reduction result from lane 63 and broadcast with	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1202	// Step 1: extract EM bits for elements 0,2,4,6 and 1,3,5,7 respectively.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1209	// Step 2: convert fp4 to fp8 using LUT	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1219	// Step 3: extract sign bits	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1223	// Step 4: assemble 4 packed fp8 values w/ sign	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1227	// Step 5: convert fp8 to fp32	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1292	// Step 1: extract EM bits for elements 0,2,4,6 and 1,3,5,7 respectively.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1300	// Step 2: extract S bit for elements 0,2,4,6 and 1,3,5,7	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1308	// Step 3: Upcast elements 0,2,4,6 to 4 16-bit elements	COMMENT
LOW	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1337	// Step 4: Upcast elements 1,3,5,7 to 4 16-bit elements	COMMENT
LOW	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1367	// Step 5: Reorder 16-bit elements to be 0,1,2,3,4,5,6,7	COMMENT
LOW	…_party/amd/lib/TritonAMDGPUToLLVM/DotOpToLLVM/MFMA.cpp	560	// Step 1: construct each kBase-element vector by	COMMENT
LOW	…_party/amd/lib/TritonAMDGPUToLLVM/DotOpToLLVM/MFMA.cpp	572	// Step 2: process rawElems based on element type	COMMENT
LOW	…_party/amd/lib/TritonAMDGPUToLLVM/DotOpToLLVM/MFMA.cpp	592	// Step 3: Insert the processed vals into the ValueTable	COMMENT
LOW	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1228	// Step 1: collect all communications between producers and consumers.	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1239	// Step 2: group channels	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1248	// Step 3: reorder producer ops and the backward slices of the producer ops.	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1251	// Step 4: find top-level ops that contain a channel, also create new ForOps	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1271	// Step 5: Create buffers. An array of buffers for each channel.	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1279	// Step 6: Lower the loads. Also add local copy ops for non-load	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1289	// Step 7: Create tokens. A set of tokens for each group of channels for	COMMENT
LOW	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1300	// Step 8: add async communication ops (ProducerAcquire etc). Also lower	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	304	// Step 1: Append accumCnts as forOp arguments.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	308	// Step 2: Add accumCnts to yieldOp.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	317	// Step 3: Create loop arguments for the new ForOp.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	326	// Step 4: Create newForOp and take the region of the original forOp.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	332	// Step 5: Replace forOp with newForOp.	COMMENT
LOW	lib/Dialect/TritonGPU/IR/LinearLayoutConversions.cpp	1047	// Step 3: Along the "register" dim, remove any all-zero bases.	COMMENT

Verbosity Indicators41 hits · 79 pts

Severity	File	Line	Snippet	Context
LOW	python/triton/runtime/jit.py	205	# The purpose of this function is to visit everything in `arguments`	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	149	# Step 1: Extract patches for each output position	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	156	# Step 2: Stack patches into im2col matrix A (M=4, K=4)	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	167	# Step 3: Reshape filter into weight matrix W (Co=1, K=4)	COMMENT
LOW⚡	python/tutorials/gluon/13-conv-im2col.py	171	# Step 4: Output = A @ W^T	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	526	// Step 1: Right shift for 8 lanes.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	529	// Step 2: Right shift for 4 lanes.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	532	// Step 3: Right shift for 2 lanes.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	535	// Step 4: Right shift for 1 lane.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	538	// Step 5: Broadcast lane 15 of each row to all the lanes of its next row.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	541	// Step 6: Broadcast lane 31 to lane 32-63.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/TargetInfo.cpp	546	// Step 7: Read the reduction result from lane 63 and broadcast with	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1202	// Step 1: extract EM bits for elements 0,2,4,6 and 1,3,5,7 respectively.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1209	// Step 2: convert fp4 to fp8 using LUT	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1219	// Step 3: extract sign bits	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1223	// Step 4: assemble 4 packed fp8 values w/ sign	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1227	// Step 5: convert fp8 to fp32	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1292	// Step 1: extract EM bits for elements 0,2,4,6 and 1,3,5,7 respectively.	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1300	// Step 2: extract S bit for elements 0,2,4,6 and 1,3,5,7	COMMENT
LOW⚡	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1308	// Step 3: Upcast elements 0,2,4,6 to 4 16-bit elements	COMMENT
LOW	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1337	// Step 4: Upcast elements 1,3,5,7 to 4 16-bit elements	COMMENT
LOW	third_party/amd/lib/TritonAMDGPUToLLVM/Utility.cpp	1367	// Step 5: Reorder 16-bit elements to be 0,1,2,3,4,5,6,7	COMMENT
LOW	…_party/amd/lib/TritonAMDGPUToLLVM/DotOpToLLVM/MFMA.cpp	560	// Step 1: construct each kBase-element vector by	COMMENT
LOW	…_party/amd/lib/TritonAMDGPUToLLVM/DotOpToLLVM/MFMA.cpp	572	// Step 2: process rawElems based on element type	COMMENT
LOW	…_party/amd/lib/TritonAMDGPUToLLVM/DotOpToLLVM/MFMA.cpp	592	// Step 3: Insert the processed vals into the ValueTable	COMMENT
LOW	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1228	// Step 1: collect all communications between producers and consumers.	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1239	// Step 2: group channels	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1248	// Step 3: reorder producer ops and the backward slices of the producer ops.	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1251	// Step 4: find top-level ops that contain a channel, also create new ForOps	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1271	// Step 5: Create buffers. An array of buffers for each channel.	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1279	// Step 6: Lower the loads. Also add local copy ops for non-load	COMMENT
LOW⚡	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1289	// Step 7: Create tokens. A set of tokens for each group of channels for	COMMENT
LOW	…/lib/Transforms/WarpSpecialization/WSCodePartition.cpp	1300	// Step 8: add async communication ops (ProducerAcquire etc). Also lower	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	304	// Step 1: Append accumCnts as forOp arguments.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	308	// Step 2: Add accumCnts to yieldOp.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	317	// Step 3: Create loop arguments for the new ForOp.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	326	// Step 4: Create newForOp and take the region of the original forOp.	COMMENT
LOW⚡	…/hopper/lib/Transforms/WarpSpecialization/WSBuffer.cpp	332	// Step 5: Replace forOp with newForOp.	COMMENT
LOW	third_party/nvidia/lib/NVGPUToLLVM/NVGPUToLLVMPass.cpp	630	// TODO: pred will have a long liverange, we need to check if this is a	COMMENT
LOW	lib/Dialect/TritonGPU/IR/LinearLayoutConversions.cpp	1047	// Step 3: Along the "register" dim, remove any all-zero bases.	COMMENT
LOW	…t/TritonInstrument/Transforms/ConcurrencySanitizer.cpp	1532	// Op is writing to the buffer, we need to check if anything else	COMMENT

Modern Structural Boilerplate75 hits · 71 pts

Severity	File	Line	Snippet	Context
LOW	python/triton/__init__.py	33	__all__ = [	CODE
LOW	python/triton/experimental/gsan/profile.py	25	__all__ = [	CODE
LOW	python/triton/experimental/gsan/__init__.py	3	__all__ = [	CODE
LOW	python/triton/experimental/gluon/__init__.py	6	__all__ = [	CODE
LOW	python/triton/experimental/gluon/_runtime.py	10	__all__ = ["GluonASTSource", "GluonJITFunction", "constexpr_function", "jit"]	CODE
LOW	python/triton/experimental/gluon/language/_core.py	57	__all__ = [	CODE
LOW	python/triton/experimental/gluon/language/_core.py	382	def _set_name(self, builder: ir.builder, name: str) -> None:	STRING
LOW	python/triton/experimental/gluon/language/amd/slice.py	8	__all__ = ["slice"]	CODE
LOW	…thon/triton/experimental/gluon/language/amd/_buffer.py	53	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
LOW	…hon/triton/experimental/gluon/language/amd/__init__.py	10	__all__ = [	CODE
LOW	python/triton/experimental/gluon/language/amd/hint.py	3	__all__ = ["sched_barrier"]	CODE
LOW	…hon/triton/experimental/gluon/language/amd/_layouts.py	9	__all__ = [	CODE
LOW	…on/experimental/gluon/language/amd/cdna5/async_copy.py	14	__all__ = ["global_to_shared", "shared_to_global", "commit_group", "wait_group", "mbarrier_arrive"]	CODE
LOW	…experimental/gluon/language/amd/cdna5/_cache_policy.py	9	__all__ = ["CachePolicy"]	CODE
LOW	…iton/experimental/gluon/language/amd/cdna5/__init__.py	17	__all__ = [	CODE
LOW	…on/triton/experimental/gluon/language/amd/cdna5/tdm.py	24	__all__ = [	CODE
LOW	…on/triton/experimental/gluon/language/amd/cdna5/tdm.py	76	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
LOW	…riton/experimental/gluon/language/amd/cdna5/cluster.py	10	__all__ = ["arrive", "wait"]	CODE
LOW	…triton/experimental/gluon/language/amd/cdna5/buffer.py	18	__all__ = ["buffer_descriptor", "buffer_descriptor_type", "make_buffer_descriptor", "load", "store"]	CODE
LOW	…iton/experimental/gluon/language/amd/cdna5/_layouts.py	9	__all__ = [	CODE
LOW	…iton/experimental/gluon/language/amd/cdna5/mbarrier.py	13	__all__ = ["MBarrierLayout", "init", "wait", "arrive"]	CODE
LOW	…on/experimental/gluon/language/amd/cdna4/async_copy.py	19	__all__ = [	CODE
LOW	…iton/experimental/gluon/language/amd/cdna4/__init__.py	17	__all__ = [	CODE
LOW	…triton/experimental/gluon/language/amd/cdna4/buffer.py	15	__all__ = [	CODE
LOW	…experimental/gluon/language/amd/cdna3/_cache_policy.py	9	__all__ = ["CachePolicy"]	CODE
LOW	…iton/experimental/gluon/language/amd/cdna3/__init__.py	20	__all__ = [	CODE
LOW	…triton/experimental/gluon/language/amd/cdna3/buffer.py	16	__all__ = [	CODE
LOW	…iton/experimental/gluon/language/amd/rdna4/__init__.py	4	__all__ = ["wmma"]	CODE
LOW	…iton/experimental/gluon/language/amd/rdna3/__init__.py	4	__all__ = ["wmma"]	CODE
LOW	…/triton/experimental/gluon/language/nvidia/__init__.py	6	__all__ = ["ampere", "blackwell", "hopper", "rubin"]	CODE
LOW	…n/experimental/gluon/language/nvidia/rubin/__init__.py	38	__all__ = [	CODE
LOW	…n/experimental/gluon/language/nvidia/rubin/mbarrier.py	11	__all__ = [	CODE
LOW	…/experimental/gluon/language/nvidia/hopper/__init__.py	11	__all__ = [	CODE
LOW	…/experimental/gluon/language/nvidia/hopper/__init__.py	94	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
LOW	…n/experimental/gluon/language/nvidia/hopper/cluster.py	5	__all__ = ["barrier"]	CODE
LOW	…riton/experimental/gluon/language/nvidia/hopper/tma.py	13	__all__ = [	CODE
LOW	…riton/experimental/gluon/language/nvidia/hopper/tma.py	122	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
LOW	…/experimental/gluon/language/nvidia/hopper/mbarrier.py	4	__all__ = [	CODE
LOW	…perimental/gluon/language/nvidia/blackwell/__init__.py	21	__all__ = [	CODE
LOW	…perimental/gluon/language/nvidia/blackwell/__init__.py	327	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
LOW	…on/experimental/gluon/language/nvidia/blackwell/tma.py	23	__all__ = [	CODE
LOW	…on/experimental/gluon/language/nvidia/blackwell/clc.py	18	__all__ = [	CODE
LOW	…on/experimental/gluon/language/nvidia/blackwell/clc.py	86	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
LOW	…xperimental/gluon/language/nvidia/ampere/async_copy.py	5	__all__ = [	CODE
LOW	…/experimental/gluon/language/nvidia/ampere/__init__.py	13	__all__ = ["CachePolicy", "FractionalEvictionPolicy", "async_copy", "mbarrier", "mma_v2"]	CODE
LOW	…/experimental/gluon/language/nvidia/ampere/mbarrier.py	8	__all__ = ["allocate_mbarrier", "arrive", "init", "invalidate", "MBarrierLayout", "wait"]	CODE
LOW	…n/triton/experimental/gluon/language/extra/__init__.py	3	__all__ = ["libdevice"]	CODE
LOW	python/triton/experimental/gluon/amd/cdna5.py	7	__all__ = ["TensorDescriptor"]	CODE
LOW	python/triton/experimental/gluon/amd/__init__.py	3	__all__ = ["cdna5", "gfx1250"]	CODE
LOW	python/triton/experimental/gluon/nvidia/blackwell.py	3	__all__ = ["TensorDescriptor"]	CODE
LOW	python/triton/experimental/gluon/nvidia/__init__.py	4	__all__ = ["hopper", "blackwell"]	CODE
LOW	python/triton/experimental/gluon/nvidia/hopper.py	7	__all__ = ["TensorDescriptor", "TensorDescriptorIm2Col"]	CODE
LOW	…riton/tools/triton_to_gluon_translator/slice_kernel.py	32	logger = logging.getLogger(__name__)	CODE
LOW	python/triton/runtime/__init__.py	7	__all__ = [	CODE
LOW	python/triton/runtime/interpreter.py	1020	def set_attr(self, obj: object, name: str, value: object) -> None:	CODE
LOW	python/triton/runtime/_allocation.py	29	def set_allocator(allocator: Allocator) -> None:	CODE
LOW	python/triton/runtime/_allocation.py	59	def set_profile_allocator(allocator: Optional[Allocator]) -> None:	CODE
LOW	python/triton/runtime/driver.py	42	def set_active(self, driver: DriverBase) -> None:	CODE
LOW	python/triton/language/__init__.py	149	__all__ = [	CODE
LOW	python/triton/language/core.py	161	def _set_name(self, builder: ir.builder, name: str) -> None:	CODE
15 more matches not shown…

Excessive Try-Catch Wrapping48 hits · 55 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	555	except Exception:	CODE
LOW	setup.py	563	except Exception:	CODE
MEDIUM	setup.py	551	def get_git_commit_hash(length=8):	CODE
MEDIUM	setup.py	559	def get_git_branch():	CODE
LOW	python/test/unit/runtime/test_cache.py	1197	except Exception as e:	STRING
LOW⚡	python/test/unit/language/test_core.py	6351	except Exception as e:	STRING
LOW	python/test/unit/language/test_decorator.py	27	except Exception as e:	CODE
LOW	python/test/backend/test_mir_stage.py	11	except Exception:	CODE
MEDIUM	python/test/backend/test_mir_stage.py	8	def is_hip():	CODE
LOW	python/triton/_compile_warmup_pool.py	313	except Exception as error:	CODE
LOW	python/triton/_compile_warmup_pool.py	509	except Exception as error:	CODE
LOW	python/triton/_compile_warmup_pool.py	515	except Exception as error:	CODE
MEDIUM	python/triton/_compile_warmup_pool.py	438	def serialize_specialization(name, signature, constants, attrs, options, key, ta	CODE
LOW	python/triton/_internal_testing.py	329	except Exception as e:	CODE
LOW	python/triton/experimental/gsan/symmetric_memory.py	174	except Exception:	CODE
LOW	python/triton/experimental/gsan/symmetric_memory.py	285	except Exception:	CODE
LOW	python/triton/experimental/gsan/symmetric_memory.py	479	except Exception:	CODE
MEDIUM	python/triton/experimental/gsan/symmetric_memory.py	282	def __del__(self) -> None:	CODE
LOW	python/triton/experimental/gsan/_utils.py	113	except Exception:	CODE
LOW	python/triton/runtime/cache.py	81	except Exception:	CODE
LOW	python/triton/runtime/cache.py	227	except Exception:	CODE
LOW	python/triton/runtime/_async_compile.py	23	except Exception:	CODE
LOW	python/triton/runtime/interpreter.py	1522	except Exception as e:	CODE
LOW	python/triton/runtime/interpreter.py	1619	except Exception:	CODE
LOW	python/triton/runtime/interpreter.py	1715	except Exception as e:	CODE
LOW	python/triton/runtime/autotuner.py	174	except Exception as e:	CODE
LOW	python/triton/compiler/compiler.py	308	except Exception as e:	CODE
LOW	python/triton/compiler/code_generator.py	1465	except Exception as e:	CODE
LOW	python/triton/compiler/code_generator.py	1516	except Exception as e:	CODE
LOW	python/triton/compiler/code_generator.py	1681	except Exception as e:	CODE
LOW	python/triton/compiler/code_generator.py	1716	except Exception as e:	CODE
LOW	python/examples/gluon/02-conv-dgrad.py	980	except Exception:	CODE
MEDIUM	python/examples/gluon/02-conv-dgrad.py	937	def _benchmark_dgrad_config(	CODE
LOW	python/examples/gluon/02-conv-wgrad.py	735	except Exception:	CODE
MEDIUM	python/examples/gluon/02-conv-wgrad.py	692	def _benchmark_wgrad_config(	CODE
LOW	third_party/proton/test/test_instrumentation.py	1170	except Exception as e:	CODE
MEDIUM	third_party/proton/test/test_instrumentation.py	1167	def run_kernel():	CODE
LOW	third_party/proton/test/test_api.py	98	except Exception as e:	CODE
LOW	third_party/proton/test/test_api.py	122	except Exception as e:	CODE
LOW	third_party/proton/test/test_api.py	135	except Exception as e:	CODE
LOW	third_party/proton/test/test_api.py	289	except Exception as e:	CODE
LOW	third_party/proton/test/test_api.py	406	except Exception as e:	CODE
LOW	third_party/proton/test/test_api.py	416	except Exception as e:	CODE
LOW	…ton/test/reproducers/cupti_graph_replay_heap_growth.py	275	except Exception:	CODE
LOW	third_party/proton/proton/proton.py	46	except Exception as e:	STRING
MEDIUM	third_party/proton/proton/proton.py	47	print("An error occurred while executing the script:")	STRING
LOW⚡	third_party/amd/python/test/test_gluon_cdna5.py	2384	except Exception as e:	CODE
LOW	third_party/nvidia/backend/driver.py	66	except Exception:	CODE

Empty Error Handlers10 hits · 50 pts

Severity	File	Line	Snippet	Context
HIGH	setup.py	82	except subprocess.CalledProcessError:	CODE
HIGH	setup.py	84	except FileNotFoundError:	CODE
HIGH	python/test/unit/language/test_annotations.py	51	except AttributeError:	CODE
HIGH	python/triton/experimental/gsan/symmetric_memory.py	285	except Exception:	CODE
HIGH	python/triton/tools/compile.py	109	except ValueError:	CODE
HIGH	python/triton/tools/compile.py	114	except ValueError:	CODE
HIGH	python/triton/language/core.py	112	except TypeError:	CODE
HIGH	third_party/proton/test/test_profile.py	1115	except ValueError:	CODE
HIGH	third_party/amd/backend/driver.py	119	except (subprocess.CalledProcessError, FileNotFoundError):	CODE
HIGH	third_party/nvidia/backend/driver.py	66	except Exception:	CODE

Self-Referential Comments16 hits · 46 pts

Severity	File	Line	Snippet	Context
MEDIUM	python/build_helpers.py	439	# Create a stable symlink that doesn't include revision	COMMENT
MEDIUM	python/test/gluon/test_core.py	5409	# Create a test matrix with known values	COMMENT
MEDIUM	python/test/gluon/test_core.py	6338	# Create a test matrix [N, M]	COMMENT
MEDIUM	python/test/gluon/test_core.py	6499	# Create a test tensor [N, M, P]	COMMENT
MEDIUM	python/triton/experimental/gluon/language/_semantic.py	669	# Create the warp specialize op.	COMMENT
MEDIUM	python/triton/experimental/gluon/language/_layouts.py	617	# Create a idendity mapping based on shape_per_cta + order	COMMENT
MEDIUM	python/triton/runtime/jit.py	414	# Create the function argument list and the dict entries for the return statement	COMMENT
MEDIUM	python/triton/runtime/jit.py	1189	# Create a bound function to support constexpr_function methods	STRING
MEDIUM	python/triton/language/core.py	1720	# Define the wrapped Triton value type.	COMMENT
MEDIUM	python/tutorials/02-fused-softmax.py	173	# Create a number of persistent programs.	COMMENT
MEDIUM	python/tutorials/01-vector-add.py	46	# Create a mask to guard memory operations against out-of-bounds accesses.	COMMENT
MEDIUM	docs/conf.py	6	# This file is execfile()d with the current directory set to its	COMMENT
MEDIUM	third_party/proton/proton/hooks/instrumentation.py	39	# Create the buffer	COMMENT
MEDIUM	third_party/proton/proton/hooks/instrumentation.py	92	# Create the appropriate mode instance	COMMENT
MEDIUM	third_party/amd/python/test/test_gluon_cdna5.py	2602	# Create a flattened view into the nD reference to unify the indexing logic over all dimensions	COMMENT
MEDIUM	third_party/amd/python/examples/gluon/mxfp_fa_cdna5.py	3151	# Create a pytest id	COMMENT

AI Slop Vocabulary16 hits · 42 pts

Severity	File	Line	Snippet	Context
MEDIUM	…iton/experimental/gluon/language/amd/cdna5/_layouts.py	106	"""Create partitioned shared memory layouts and WMMA layout for a CDNA5 GEMM in order to avoid LDS partition conflic	STRING
MEDIUM	…riton/tools/triton_to_gluon_translator/slice_kernel.py	135	# HACK: An AST rewrite would be more robust, but this works for now.	COMMENT
MEDIUM	python/triton/runtime/cache.py	117	# use temp dir to be robust against program interruptions	COMMENT
LOW	python/triton/runtime/interpreter.py	1616	# e.g., dynamically generated functions, we cannot rewrite it so just return the original function	COMMENT
MEDIUM	python/triton_kernels/bench/bench_dense_matmul.py	3	# This is not meant to be a comprehensive benchmark of triton_kernels.	COMMENT
MEDIUM	python/triton_kernels/bench/bench_dense_matmul.py	135	# even for the fp16 benchmark path. Match that when reproducing the OpenAI harness.	COMMENT
MEDIUM	python/tutorials/09-persistent-matmul.py	419	# Enable warp specialization to leverage async warp scheduling in the GPU.	COMMENT
MEDIUM	python/tutorials/10-block-scaled-matmul.py	405	# With 256 elements we better utilize the L2 and don't require the TMA	COMMENT
MEDIUM	python/tutorials/gluon/11-tcgen05-mma-scaled.py	654	# elements. With 256 bytes along the inner dimension, we better utilize the	COMMENT
MEDIUM	python/tutorials/gluon/08-warp-specialization.py	369	# specialization is especially powerful when used to build intricate schedules	COMMENT
MEDIUM	python/tutorials/gluon/02-layouts.py	868	# Linear layouts are extremely poweful, and can be used in conjunction with	COMMENT
MEDIUM	python/tutorials/gluon/07-persistence.py	841	# work in conjunction with the GPU, getting the best of both worlds. This is	COMMENT
LOW	third_party/amd/python/test/test_gluon_cdna5.py	3170	# In practice, since we know that phase is initialized to 0, we can just set count = NUM_WARPS * WARP_SIZE and call	COMMENT
LOW	third_party/amd/python/test/test_gluon_cdna5.py	3231	# In practice, since we know that phase is initialized to 0, we can just set count = NUM_WARPS and call directly ttg	COMMENT
MEDIUM	…lib/TritonAMDGPUTransforms/Pipeliner/ScheduleLoops.cpp	210	/// which is not coalesced. To better utilize global memory bandwidth,	COMMENT
MEDIUM	…party/amd/lib/TritonAMDGPUToLLVM/LoadStoreOpToLLVM.cpp	2522	// In the case of unpaired f16 elements utilize dpp instructions to	COMMENT

AI Structural Patterns38 hits · 36 pts

Severity	File	Line	Context
LOW	python/test/unit/language/test_core.py	226	CODE
LOW	python/triton/testing.py	43	CODE
LOW	python/triton/experimental/gluon/_runtime.py	81	CODE
LOW	python/triton/tools/ragged_tma.py	76	CODE
LOW	…/tools/triton_to_gluon_translator/blackwell_helpers.py	93	CODE
LOW	…/tools/triton_to_gluon_translator/blackwell_helpers.py	205	CODE
LOW	…/tools/triton_to_gluon_translator/blackwell_helpers.py	304	CODE
LOW	…triton/tools/triton_to_gluon_translator/amd_helpers.py	302	CODE
LOW	…ton/tools/triton_to_gluon_translator/nvidia_helpers.py	53	CODE
LOW	…ton/tools/triton_to_gluon_translator/nvidia_helpers.py	133	CODE
LOW	…/triton/tools/triton_to_gluon_translator/translator.py	257	CODE
LOW	python/triton/runtime/jit.py	991	CODE
LOW	python/triton/runtime/autotuner.py	424	CODE
LOW	python/triton/runtime/autotuner.py	21	CODE
LOW	python/triton/language/standard.py	448	CODE
LOW	python/triton/language/core.py	3697	CODE
LOW	python/triton/compiler/code_generator.py	296	CODE
LOW	python/examples/gluon/04-2cta-block-scale-matmul.py	818	CODE
LOW	python/triton_kernels/triton_kernels/matmul.py	248	CODE
LOW	python/triton_kernels/triton_kernels/matmul.py	901	CODE
LOW	python/triton_kernels/triton_kernels/reduce.py	669	CODE
LOW	python/triton_kernels/triton_kernels/reduce.py	1150	CODE
LOW	python/triton_kernels/triton_kernels/reduce.py	1188	CODE
LOW	python/triton_kernels/triton_kernels/testing.py	295	CODE
LOW	python/triton_kernels/triton_kernels/swiglu.py	100	CODE
LOW	python/triton_kernels/triton_kernels/swiglu.py	82	CODE
LOW	…n_kernels/tensor_details/layout_details/cdna4_scale.py	77	CODE
LOW	…_kernels/tensor_details/layout_details/hopper_value.py	288	CODE
LOW	…_kernels/tensor_details/layout_details/hopper_value.py	252	CODE
LOW	…rnels/tensor_details/layout_details/blackwell_scale.py	222	CODE
LOW	…_kernels/tensor_details/layout_details/hopper_scale.py	147	CODE
LOW	…ton_kernels/triton_kernels/matmul_details/_p_matmul.py	84	CODE
LOW	…riton_kernels/triton_kernels/matmul_details/_matmul.py	38	CODE
LOW	python/tutorials/02-fused-softmax.py	59	CODE
LOW	python/tutorials/gluon/14-multicta.py	1066	CODE
LOW	…ird_party/amd/python/examples/gluon/mxfp_gemm_cdna5.py	132	CODE
LOW	third_party/amd/python/examples/gluon/moe_cdna5.py	1202	CODE
LOW	third_party/amd/python/examples/gluon/mxfp_fa_cdna5.py	3006	CODE

Overly Generic Function Names20 hits · 21 pts

Severity	File	Line	Snippet	Context
LOW⚡	python/test/unit/tools/test_slice_kernel.py	784	def helper() -> None:	CODE
LOW⚡	python/test/unit/tools/test_slice_kernel.py	800	def helper() -> int:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	165	def helper() -> int:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	192	def helper() -> int:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	213	def helper() -> int:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	240	def helper() -> int:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	501	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	516	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	536	def helper():	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	551	def helper():	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	574	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	593	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	642	def helper():	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	657	def helper():	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	684	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	703	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	726	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	745	def helper() -> None:	CODE
LOW	python/test/unit/tools/test_slice_kernel.py	768	def helper() -> None:	CODE
LOW	python/test/unit/runtime/test_driver.py	290	def helper(value):	CODE

Cross-File Repetition3 hits · 15 pts

Severity	File	Snippet	Context
HIGH	…iton/experimental/gluon/language/amd/cdna5/__init__.py	computes matrix multiplication ``a * b + acc`` using an amd wmma instruction. args: a (tensor): the operand a to be mult	STRING
HIGH	…iton/experimental/gluon/language/amd/rdna4/__init__.py	computes matrix multiplication ``a * b + acc`` using an amd wmma instruction. args: a (tensor): the operand a to be mult	STRING
HIGH	…iton/experimental/gluon/language/amd/rdna3/__init__.py	computes matrix multiplication ``a * b + acc`` using an amd wmma instruction. args: a (tensor): the operand a to be mult	STRING

Redundant / Tautological Comments10 hits · 13 pts

Severity	File	Line	Snippet	Context
LOW	python/build_helpers.py	131	# Set timeout to 300 seconds to prevent the request from hanging forever.	COMMENT
LOW	python/triton/compiler/code_generator.py	151	# Check if the given syntax node has an "early" return	COMMENT
LOW	utils/generate-test-checks.py	177	# Check if an existing variable exists for this name.	STRING
LOW	utils/generate-test-checks.py	320	# Check if this is an attribute definition and process it	STRING
LOW	third_party/proton/proton/hooks/hook.py	93	# Check if any other sessions rely on this hook	COMMENT
LOW	third_party/amd/python/test/address_sanitizer_helper.py	23	#Set access to go out of bounds for ASAN test	COMMENT
LOW	third_party/amd/backend/driver.py	290	# Check if cooperative groups are supported on the device.	COMMENT
LOW	.github/workflows/integration-tests-nvidia.yml	45	# Check if files exist before proceeding	COMMENT
LOW	.github/workflows/integration-tests-amd.yml	91	# Check if files exist before proceeding	COMMENT
LOW	.github/workflows/build-macos.yml	38	# Check if files exist before proceeding	COMMENT

Cross-Language Confusion3 hits · 12 pts

Severity	File	Line	Snippet	Context
HIGH	python/test/unit/tools/test_aot.py	147	while (fscanf(file, "%hd,", &buffer[index]) != EOF && index < size) {	CODE
HIGH	python/triton/experimental/gsan/symmetric_memory.py	431	stack.push(listener)	CODE
HIGH	python/triton/tools/link.py	218	src += " if (algo_id < 0 \|\| algo_id >= num_algos)\n"	STRING

Slop Phrases4 hits · 8 pts

Severity	File	Line	Snippet	Context
LOW	python/test/unit/language/test_core.py	7889	# reuse the temp variable and make sure to check that it isn't creating incorrect IR.	STRING
LOW	python/tutorials/01-vector-add.py	74	# - Don't forget to pass meta-parameters as keywords arguments.	COMMENT
MEDIUM	python/tutorials/gluon/14-multicta.py	357	# It's worth noting that the pattern changes a bit once the TMA has to wait on the tcgen05_mma	COMMENT
MEDIUM	python/tutorials/gluon/02-layouts.py	839	# require reordering registers within a thread (which is free), you can use	COMMENT

AI Response Leakage1 hit · 8 pts

Severity	File	Line	Snippet	Context
HIGH	python/tutorials/11-programmatic-dependent-launch.py	26	# In this example	COMMENT

Fake / Example Data11 hits · 6 pts

Severity	File	Line	Snippet	Context
LOW⚡	python/test/unit/language/test_line_info.py	584	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	344	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	364	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	383	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	396	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	416	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	442	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	466	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	493	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/test/unit/language/test_line_info.py	605	run_filecheck("placeholder", h.asm["ttir"], check_template)	STRING
LOW	python/triton/_filecheck.py	87	run_filecheck("placeholder", mlir_module.str_nodebug(), check_template)	CODE

Docstring Block Structure1 hit · 5 pts

Severity	File	Line	Snippet	Context
HIGH	…on/triton/experimental/gluon/language/amd/cdna5/tdm.py	172	Update selected fields of a TDM descriptor; return a new descriptor SSA value. Each parameter is independently opti	STRING

TODO Padding3 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW	third_party/amd/backend/include/hip/hip_runtime_api.h	2500	// TODO: implement IPC apis	COMMENT
LOW	lib/Analysis/Alias.cpp	85	// TODO: implement	COMMENT
LOW	lib/Analysis/Alias.cpp	91	// TODO: implement	COMMENT

Dead Code2 hits · 4 pts

Severity	File	Line	Snippet	Context
MEDIUM	python/test/unit/language/test_compile_errors.py	172		CODE
MEDIUM	python/triton_kernels/triton_kernels/tensor.py	98		CODE

Example Usage Blocks1 hit · 2 pts

Severity	File	Line	Snippet	Context
LOW	bin/triton-tensor-layout.cpp	23	// Example usage:	COMMENT

triton-lang/triton

Open-Source Code Quality, Security and Technical Debt Review