MCPcopy Create free account

hub / github.com/tile-ai/tilelang / functions

Functions7,254 in github.com/tile-ai/tilelang

↓ 1 callersFunctionregion_is_same
(a, b)
tilelang/carver/roller/shape_inference/tir.py:305
↓ 1 callersFunctionregister_let_value
Record the value behind a flat tirx Bind. The legacy LetFrame path updated FrameStack through __enter__/__exit__. tirx emits flat Bind statem
tilelang/language/frame.py:117
↓ 1 callersFunctionregression_all
Run all example benchmark drivers and print a consolidated table. Intended usage (CI): `python -c "import tilelang.testing.perf_regression as pr;
maint/scripts/regression_all.py:73
↓ 1 callersMethodreplace_weight_param_with_qweight
(self)
examples/bitnet-1.58b/utils_quant.py:92
↓ 1 callersMethodreset
(self, task_manager: TaskManager)
tilelang/autodd.py:1154
↓ 1 callersFunctionreset_recorder
Begin a fresh recording window on this thread.
tilelang/contrib/hip_resource_info.py:64
↓ 1 callersFunctionreshape_fragment_test
(N, M, dtype)
testing/python/language/test_tilelang_language_reshape.py:122
↓ 1 callersFunctionreshape_layout_transform_shared
(N, M, dtype)
testing/python/language/test_tilelang_language_reshape.py:162
↓ 1 callersFunctionreshape_shape_mismatch_test
(N, M, dtype)
testing/python/language/test_tilelang_language_reshape.py:245
↓ 1 callersFunctionreshape_test
(N, M, dtype)
testing/python/language/test_tilelang_language_reshape.py:8
↓ 1 callersFunctionreshape_test_smem_1d_2_2d
(N, M, dtype)
testing/python/language/test_tilelang_language_reshape.py:44
↓ 1 callersFunctionreshape_test_smem_2d_2_1d
(N, M, dtype)
testing/python/language/test_tilelang_language_reshape.py:83
↓ 1 callersFunctionresolve_gemm_sp_impl
Resolve the registered GEMM_SP implementation class for a target.
tilelang/tileop/gemm_sp/registry.py:38
↓ 1 callersMethodrewrite
(self, node: ast.AST, parent: ast.AST, field: str, inside_list: bool)
tilelang/autodd.py:96
↓ 1 callersFunctionrocm_link
Link relocatable ELF object to shared ELF object using lld Parameters ---------- in_file : str Input file name (relocatable ELF o
tilelang/contrib/rocm.py:66
↓ 1 callersFunctionruff_fix_code
(code_string: str, fix_lint: bool = True, format_code: bool = True)
tilelang/autodd.py:652
↓ 1 callersFunctionrun_abs_test
Test that abs correctly maps to fabs (not __fabsf) in generated CUDA code
testing/python/fastmath/test_mathops_fastmath.py:165
↓ 1 callersFunctionrun_abs_test
Test that abs correctly maps to fabs (not __fabsf) in generated CUDA code
testing/python/math/test_math_fast_math.py:165
↓ 1 callersFunctionrun_alloc_global
( N, block_N, dtype, )
testing/python/language/test_tilelang_language_alloc.py:184
↓ 1 callersFunctionrun_alloc_global_eagerjit
( N, block_N, dtype, )
testing/python/language/test_tilelang_language_alloc.py:224
↓ 1 callersFunctionrun_alloc_multi_vars_with_initializer
( N, block_N, dtype, )
testing/python/language/test_tilelang_language_alloc.py:147
↓ 1 callersFunctionrun_alloc_var
( N, block_N, dtype, min=None, max=None, )
testing/python/language/test_tilelang_language_alloc.py:26
↓ 1 callersFunctionrun_alloc_var_add
( N, block_N, dtype, )
testing/python/language/test_tilelang_language_alloc.py:68
↓ 1 callersFunctionrun_alloc_var_with_initializer
( N, block_N, dtype, init_value, )
testing/python/language/test_tilelang_language_alloc.py:106
↓ 1 callersMethodrun_async
(self, task_manager: TaskManager)
tilelang/autodd.py:1225
↓ 1 callersFunctionrun_atomic_add
(K, M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:25
↓ 1 callersFunctionrun_atomic_add_auto_vectorized
(K, M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:218
↓ 1 callersFunctionrun_atomic_add_complicated_parallel
(K, M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:257
↓ 1 callersFunctionrun_atomic_addx4
(M, N, block_M, block_N)
testing/python/language/test_tilelang_language_atomic.py:140
↓ 1 callersFunctionrun_atomic_load_store
(M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:498
↓ 1 callersFunctionrun_atomic_max
(K, M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:432
↓ 1 callersFunctionrun_atomic_memory_order
(K, M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:51
↓ 1 callersFunctionrun_atomic_min
(K, M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:465
↓ 1 callersFunctionrun_atomic_return_prev
(M, N, block_M, block_N, dtype=T.float32)
testing/python/language/test_tilelang_language_atomic.py:173
↓ 1 callersFunctionrun_auto_tma_store_copy
()
testing/python/language/test_tilelang_language_tma_store.py:138
↓ 1 callersFunctionrun_benchmark_suite
Run a suite of benchmarks with different configurations.
examples/deepseek_nsa/benchmark/benchmark_nsa_fwd.py:788
↓ 1 callersFunctionrun_block_sparse_matmul_global
(M=1024, N=1024, K=1024, sparsity=0.5, condition_dim=2)
testing/python/language/test_tilelang_language_any_of.py:165
↓ 1 callersFunctionrun_block_sparse_matmul_global
(M=1024, N=1024, K=1024, sparsity=0.5, condition_dim=2)
testing/python/language/test_tilelang_language_all_of.py:165
↓ 1 callersFunctionrun_block_sparse_matmul_local
(M=1024, N=1024, K=1024, sparsity=0.5, condition_dim=2)
testing/python/language/test_tilelang_language_any_of.py:251
↓ 1 callersFunctionrun_block_sparse_matmul_local
(M=1024, N=1024, K=1024, sparsity=0.5, condition_dim=2)
testing/python/language/test_tilelang_language_all_of.py:251
↓ 1 callersFunctionrun_block_sparse_matmul_shared
(M=1024, N=1024, K=1024, sparsity=0.5, condition_dim=2)
testing/python/language/test_tilelang_language_any_of.py:206
↓ 1 callersFunctionrun_block_sparse_matmul_shared
(M=1024, N=1024, K=1024, sparsity=0.5, condition_dim=2)
testing/python/language/test_tilelang_language_all_of.py:206
↓ 1 callersFunctionrun_blockscaled_mxfp4_x_mxfp8
(M: int = 128, N: int = 256, K: int = 128)
maint/gemm/correctness_evaluation_tcgen05_fp4_unpacked.py:333
↓ 1 callersFunctionrun_chain_equal
(N=128, block_size=64, dtype=T.float32)
testing/python/language/test_tilelang_language_chain_equal.py:25
↓ 1 callersFunctionrun_cutedsl_kernel_do_bench
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M, block_N, block_K, num_stages=3, num
testing/python/jit/test_tilelang_jit_cutedsl.py:237
↓ 1 callersFunctionrun_cutedsl_kernel_multi_stream
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M, block_N, block_K, num_stages=3, num
testing/python/jit/test_tilelang_jit_cutedsl.py:277
↓ 1 callersFunctionrun_cython_cluster_launch
()
testing/python/language/test_tilelang_language_cluster.py:74
↓ 1 callersFunctionrun_cython_dynamic_shape_with_out_idx
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M, block_N, block_K, num_stages=3, num
testing/python/jit/test_tilelang_jit_gemm_cython.py:309
↓ 1 callersFunctionrun_cython_kernel_do_bench
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M, block_N, block_K, num_stages=3, num
testing/python/jit/test_tilelang_jit_gemm_cython.py:177
↓ 1 callersFunctionrun_cython_kernel_multi_stream
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M, block_N, block_K, num_stages=3, num
testing/python/jit/test_tilelang_jit_gemm_cython.py:215
↓ 1 callersFunctionrun_fastmath_mathop_test
Test fastmath mathops to ensure they generate fastmath CUDA code (with __ prefix).
testing/python/fastmath/test_mathops_fastmath.py:201
↓ 1 callersFunctionrun_fastmath_mathop_test
Test fastmath mathops to ensure they generate fastmath CUDA code (with __ prefix).
testing/python/math/test_math_fast_math.py:201
↓ 1 callersFunctionrun_fill_static
(n: int)
testing/python/language/test_tilelang_language_int64.py:113
↓ 1 callersFunctionrun_fill_symbolic
(n: int)
testing/python/language/test_tilelang_language_int64.py:84
↓ 1 callersFunctionrun_fp4_tma_copy_roundtrip
()
testing/python/language/test_tilelang_language_tma_copy.py:376
↓ 1 callersFunctionrun_fp4_tma_copy_unpacked_smem_load
()
testing/python/language/test_tilelang_language_tma_copy.py:447
↓ 1 callersFunctionrun_fp4_unpacked_x_fp8
(M: int = 128, N: int = 256, K: int = 128)
maint/gemm/correctness_evaluation_tcgen05_fp4_unpacked.py:298
↓ 1 callersFunctionrun_full_shape_tma_store_1d_codegen
()
testing/python/language/test_tilelang_language_tma_store.py:163
↓ 1 callersFunctionrun_gather_scatter
(N=64, K=64, K_box=64)
testing/python/language/test_tilelang_language_tma_gather_scatter.py:59
↓ 1 callersFunctionrun_gather_scatter_swizzled
(N, K, K_box, swizzle_kind)
testing/python/language/test_tilelang_language_tma_gather_scatter.py:148
↓ 1 callersFunctionrun_gemm
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
maint/gemm/correctness_evaluation_tcgen05.py:94
↓ 1 callersFunctionrun_gemm
( M, N, K, in_dtype, out_dtype, accum_dtype, block_M, block_N, block_K,
maint/gemm/correctness_evaluation_tcgen05_2cta.py:93
↓ 1 callersFunctionrun_gemm
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_callback.py:55
↓ 1 callersFunctionrun_gemm_batched
( batch, M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, b
testing/python/kernel/test_tilelang_kernel_gemm_batched.py:55
↓ 1 callersFunctionrun_gemm_jit_kernel
(M, N, K, block_M, block_N, block_K)
testing/python/issue/test_tilelang_issue_1697.py:28
↓ 1 callersFunctionrun_gemm_jit_kernel
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_cutedsl.py:162
↓ 1 callersFunctionrun_gemm_jit_kernel
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_tvm_ffi.py:101
↓ 1 callersFunctionrun_gemm_jit_kernel
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_gemm_cython.py:103
↓ 1 callersFunctionrun_gemm_jit_kernel
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_callback.py:171
↓ 1 callersFunctionrun_gemm_jit_kernel
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_nvrtc.py:101
↓ 1 callersFunctionrun_gemm_kernel_jit
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/jit/test_tilelang_jit_gemm.py:56
↓ 1 callersFunctionrun_gemm_mixed_pp
()
testing/python/analysis/test_tilelang_nested_loop_checker.py:451
↓ 1 callersFunctionrun_gemm_nested_pipelines
( extra_pipeline_repeats, )
testing/python/analysis/test_tilelang_nested_loop_checker.py:139
↓ 1 callersFunctionrun_gemm_rr
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/tilelibrary/test_tilelang_tilelibrary_gemm_sp.py:594
↓ 1 callersFunctionrun_gemm_rr_false_false
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:499
↓ 1 callersFunctionrun_gemm_rr_false_true
(m, n, k, in_dtype, out_dtype, accum_dtype, num_threads=128)
maint/gemm/correctness_evaluation.py:495
↓ 1 callersFunctionrun_gemm_rr_true_false
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:503
↓ 1 callersFunctionrun_gemm_rr_true_true
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:507
↓ 1 callersFunctionrun_gemm_rs
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/tilelibrary/test_tilelang_tilelibrary_gemm_sp.py:253
↓ 1 callersFunctionrun_gemm_rs
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/kernel/test_tilelang_kernel_gemm.py:482
↓ 1 callersFunctionrun_gemm_rs_false_false
(m, n, k)
maint/gemm/correctness_evaluation_sm70.py:245
↓ 1 callersFunctionrun_gemm_rs_false_false
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:467
↓ 1 callersFunctionrun_gemm_rs_false_true
(m, n, k, in_dtype, out_dtype, accum_dtype)
maint/gemm/correctness_evaluation_sm70.py:241
↓ 1 callersFunctionrun_gemm_rs_false_true
(m, n, k, in_dtype, out_dtype, accum_dtype, num_threads=128)
maint/gemm/correctness_evaluation.py:463
↓ 1 callersFunctionrun_gemm_rs_true_false
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:471
↓ 1 callersFunctionrun_gemm_rs_true_true
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:475
↓ 1 callersFunctionrun_gemm_sr
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/tilelibrary/test_tilelang_tilelibrary_gemm_sp.py:421
↓ 1 callersFunctionrun_gemm_sr
( M, N, K, trans_A, trans_B, in_dtype, out_dtype, dtypeAccum, block_M,
testing/python/kernel/test_tilelang_kernel_gemm.py:362
↓ 1 callersFunctionrun_gemm_sr_false_false
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:483
↓ 1 callersFunctionrun_gemm_sr_false_true
(m, n, k, in_dtype, out_dtype, accum_dtype, num_threads=128)
maint/gemm/correctness_evaluation.py:479
↓ 1 callersFunctionrun_gemm_sr_true_false
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:487
↓ 1 callersFunctionrun_gemm_sr_true_true
(m, n, k, num_threads=128)
maint/gemm/correctness_evaluation.py:491
↓ 1 callersFunctionrun_gemm_tiled_op_with_parallel
()
testing/python/analysis/test_tilelang_nested_loop_checker.py:570
↓ 1 callersFunctionrun_gemm_with_stride_ss
(M: int, N: int, K: int, block_M: int, block_N: int, block_K: int)
testing/python/kernel/test_tilelang_kernel_gemm_with_stride.py:46
↓ 1 callersFunctionrun_general_matmul_emit_configs
(M, N, K, topk: int = 20)
testing/python/carver/test_tilelang_carver_generate_hints.py:9
↓ 1 callersFunctionrun_general_matmul_matmul_emit_configs
(M, N, K, topk: int = 20)
testing/python/carver/test_tilelang_carver_generate_hints.py:57
↓ 1 callersMethodrun_hopper
Perform all-reduce on Hopper architecture using bar.sync. Based on tl::AllReduce<...>::run_hopper from reduce.h S
tilelang/contrib/cutedsl/reduce.py:607
↓ 1 callersFunctionrun_kernel
(config, *args, **kwargs)
examples/autodd/tilelang_minimized_expected.py:39
↓ 1 callersFunctionrun_kernel
Run kernel - includes extra redundant logic
examples/autodd/tilelang_buggy.py:166
↓ 1 callersFunctionrun_kernel_only
()
examples/deepseek_mhc/example_mhc_pre.py:446
← previousnext →2,901–3,000 of 7,254, ranked by callers