MCPcopy Create free account

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

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

Methodlower
( self, layout_map: dict, target: Target, thread_bounds: Range, thread_var: tir.Var, mbar_phase_expr:
tilelang/metal/op/gemm/gemm_metal.py:23
Methodlower
( self, layout_map: dict, target: Target, thread_bounds: Range, thread
tilelang/cuda/op/gemm/gemm_mma_sm70.py:55
Methodlower
Lower the GEMM tile-op into a TIR prim_func containing TCGEN5MMA calls.
tilelang/cuda/op/gemm/gemm_tcgen05.py:123
Methodlower
( self, layout_map: dict, target: Target, thread_bounds: Range, thread
tilelang/cuda/op/gemm/gemm_wgmma.py:87
Methodlower
( self, layout_map: dict, target: Target, thread_bounds: Range, thread
tilelang/cuda/op/gemm_sp/gemm_sp_wgmma.py:73
Methodlower
(self, layout_map: dict, target: Target, thread_bounds: Range, thread_var: tirx.Var)
tilelang/cuda/op/gemm_sp/gemm_sp_mma.py:62
Functionmacro
Decorator that converts a Python function into a TileLang macro. TileLang macro is very similar to PrimFunc, it can be used in prim_func or a
tilelang/language/eager/builder.py:820
Functionmacro
Decorator for macro definitions. Parameters ---------- hygienic: bool Specifies whether the macro is hygienic or not. A m
tilelang/language/parser/entry.py:105
Functionmacro
Decorator for macro definitions. Parameters ---------- hygienic: bool Specifies whether the macro is hygienic or not. A m
tilelang/language/tir/entry.py:64
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
maint/gemm/correctness_evaluation_tcgen05.py:29
Functionmain
( A: T.Tensor((M, K), in_dtype), B: T.Tensor((K, N), in_dtype), C: T.Tensor((M, N), ou
maint/gemm/correctness_evaluation_tcgen05_2cta.py:22
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
maint/gemm/correctness_evaluation_sm70.py:29
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
maint/gemm/correctness_evaluation.py:30
Functionmain
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), C: T.Tensor((M, N), dtype),
maint/host_checks/common.py:8
Functionmain
( cb: T.Tensor((batch, nchunks, ngroups, chunk_size, chunk_size), dtype), # type: ignore x: T
benchmark/mamba2/benchmark_mamba_chunk_scan.py:211
Functionmain
The compiled TVM function for block-level matrix multiplication. - We divide the entire (M, N) domain into blocks of shape
benchmark/matmul/benchmark_matmul_sp.py:164
Functionmain
The compiled TVM function for block-level matrix multiplication. - We divide the entire (M, N) domain into blocks of shape
benchmark/matmul/benchmark_matmul.py:178
Functionmain
( Q: T.Tensor(shape, dtype), K: T.Tensor(shape, dtype), V: T.Tensor(shape,
benchmark/blocksparse_attention/benchmark_tilelang_block_sparse_fmha.py:121
Functionmain
The compiled TVM function for block-level matrix multiplication. - We divide the entire (M, N) domain into blocks of shape
benchmark/matmul_fp8/benchmark_matmul.py:165
Functionmain
()
examples/flash_decoding/example_gqa_decode_varlen_logits.py:286
Functionmain
( X: T.Tensor((N, D), in_dtype), gamma: T.Tensor((D,), in_dtype), beta: T.Tensor((D,),
examples/norm/layernorm.py:11
Functionmain
GPU kernel entry that performs a blocked, pipelined matrix multiplication A @ B.T writing into C. This kernel: - Loads tiles
examples/bitnet-1.58b/kernel_benchmark/tilelang_bitnet_158_int8xint2_prefill.py:185
Functionmain
( A: T.Buffer(A_shape, in_dtype), B: T.Buffer(B_shape, in_dtype), C: T.Buffer((M, N),
examples/bitnet-1.58b/kernel_benchmark/tl_int8xint8.py:104
Functionmain
()
examples/deepseek_mhc/example_mhc_post.py:131
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
examples/gemm_fp8/example_tilelang_gemm_amd_fp8_preshuffle.py:103
Functionmain
( a: T.Buffer((B, M, N), T.bfloat16), a_out: T.Buffer((B, M, N), T.float32), )
testing/python/layout/test_tilelang_layout_fused_replicate.py:16
Functionmain
(A: T.Tensor((128, 4), T.float), B: T.Tensor((128, 4), T.float))
testing/python/layout/test_tilelang_layout_inference.py:10
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/components/test_tilelang_pass_config_disable_warp_specialized.py:27
Functionmain
( A_sparse: T.Tensor(A_sparse_shape, in_dtype), E: T.Tensor(E_shape, metadata_dtype),
testing/python/utils/test_compress_utils.py:37
Functionmain
( A: T.Tensor((N,), dtype), B: T.Tensor((N,), dtype), C: T.Tensor((N,), dtype), )
testing/python/language/test_tilelang_language_chain_equal.py:12
Functionmain
( A: T.Tensor((128, 128), "float32"), B: T.Tensor((128, 128), "float32"), C: T.Tensor(
testing/python/language/test_tilelang_language_min_blocks_per_sm.py:11
Functionmain
( A: T.Tensor((M, K), in_dtype), B: T.Tensor((N, K), in_dtype), C: T.Tensor((M, N), ou
testing/python/language/test_tilelang_language_atom_mma.py:81
Functionmain
( A: T.Tensor((N,), dtype), B: T.Tensor((N // M, M), dtype), )
testing/python/language/test_tilelang_language_reshape.py:10
Functionmain
(x: T.Tensor[n, dtype])
testing/python/language/test_tilelang_language_int64.py:73
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((N, M), dtype), )
testing/python/language/test_tilelang_language_transpose.py:21
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M, N), dtype), )
testing/python/language/test_tilelang_language_ternary.py:12
Functionmain
(A: T.Tensor((N,), T.float32), l: T.int32, r: T.int32)
testing/python/language/test_tilelang_language_assume.py:12
Functionmain
( A: T.Tensor[(M,), dtype_A], # noqa: F821 B: T.Tensor[(M,), dtype_B], # noqa: F821 )
testing/python/language/test_tilelang_language_vectorized_cast.py:12
Functionmain
(A: T.Tensor((128), T.int32))
testing/python/language/test_tilelang_language_cluster.py:35
Functionmain
( A: T.Tensor((64,), T.uint8), B: T.Tensor((64,), T.uint8), )
testing/python/language/test_tilelang_language_access_ptr_codegen.py:13
Functionmain
( A: T.Tensor[(128, 8), T.float32], B: T.Tensor[(128, 8), T.float32], )
testing/python/language/test_tilelang_language_select.py:9
Functionmain
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), BlockMask: T.Tensor(block_ma
testing/python/language/test_tilelang_language_any_of.py:40
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M * N), dtype), )
testing/python/language/test_tilelang_language_composable_index.py:11
Functionmain
(A: T.Tensor((32,), dtype))
testing/python/language/test_tilelang_language_infinity.py:10
Functionmain
( A: T.Tensor((128, 128), T.bfloat16), B: T.Tensor((128, 128), T.bfloat16), D: T.Tenso
testing/python/language/test_tilelang_language_tcgen05_gemm.py:11
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/language/test_tilelang_language_tma_store.py:36
Functionmain
( A: T.Tensor((M, N), src_dtype), B: T.Tensor((M, N), dst_dtype), )
testing/python/language/test_tilelang_language_copy.py:13
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/language/test_tilelang_language_pipeline.py:28
Functionmain
( A: T.Tensor[(1,), T.int32], B: T.Tensor[(1,), T.int32], )
testing/python/language/test_tilelang_language_augassign.py:24
Functionmain
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), dtype),
testing/python/language/test_tilelang_language_clear.py:9
Functionmain
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), dtype),
testing/python/language/test_tilelang_language_alias.py:9
Functionmain
( a_ptr: T.ptr, b_ptr: T.ptr, c_ptr: T.ptr, m: T.int32, n: T.int32,
testing/python/language/test_tilelang_language_ptr.py:11
Functionmain
( A: T.Tensor((64, 16), T.float16), B: T.Tensor((16, 64), T.float16), D: T.Tensor((64,
testing/python/language/test_tilelang_language_wgmma_gemm.py:11
Functionmain
( B: T.Tensor((32,), "int32"), )
testing/python/language/test_tilelang_language_warp_vote.py:32
Functionmain
( x: T.Tensor((n,), T.float32), y: T.Tensor((n,), T.int32), )
testing/python/language/test_tilelang_language_isfinite_codegen.py:12
Functionmain
( A: T.Tensor[(M_val,), "float32"], # noqa: F821 B: T.Tensor[(M_val,), "float8_e4m3fn
testing/python/language/test_tilelang_cast_rounding.py:74
Functionmain
( A: T.Tensor((N,), dtype), B: T.Tensor((N,), dtype), )
testing/python/language/test_tilelang_language_alloc.py:12
Functionmain
( A: T.Tensor((M, K), dtype), B: T.Tensor((K, N), dtype), BlockMask: T.Tensor(block_ma
testing/python/language/test_tilelang_language_all_of.py:40
Functionmain
( A: T.StridedTensor[(N, M), (1, stride_A), T.float32], # noqa: F821 B: T.StridedTensor[(N, M
testing/python/language/test_tilelang_language_vectorize.py:12
Functionmain
( A: T.Tensor((length,), dtype), B: T.Tensor((length,), dtype), )
testing/python/language/test_tilelang_language_parallel.py:13
Functionmain
(x: T.Tensor((32), dtype))
testing/python/language/test_tilelang_language_warp_reduce.py:13
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M, N), dtype), )
testing/python/language/test_tilelang_language_mask_op.py:10
Functionmain
( A: T.Tensor((N,), T.float32), B: T.Tensor((N,), T.float32), )
testing/python/language/test_tilelang_language_source_kernel.py:28
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M, N), dtype), )
testing/python/language/test_tilelang_language_annotate_safe_value.py:11
Functionmain
(A: T.Tensor((M, N), dtype), B: T.Tensor((M,), dtype))
testing/python/language/test_tilelang_language_reduce.py:103
Functionmain
( Src: T.Tensor((N, K), in_dtype), Idx: T.Tensor((4,), "int32"), Dst: T.Tensor((N, K),
testing/python/language/test_tilelang_language_tma_gather_scatter.py:28
Functionmain
Trigger dependent PDL launches after writing the output tensor.
testing/python/language/test_tilelang_language_pdl.py:18
Functionmain
( x: T.Tensor((N,), T.float32), y: T.Tensor((N,), T.float32), )
testing/python/language/test_tilelang_language_intrinsics_codegen.py:11
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M, N), dtype), BlockMask: T.Tensor(block_ma
testing/python/language/test_tilelang_language_let_layout.py:31
Functionmain
()
testing/python/language/test_tilelang_language_float4_e2m1_unpacked_dtype.py:25
Functionmain
( A: T.Tensor((1,), "int32"), B: T.Tensor((1,), "int32"), )
testing/python/language/test_tilelang_language_warp_sync.py:11
Functionmain
(A_ptr: T.handle)
testing/python/language/test_tilelang_language_unroll.py:9
Functionmain
( A: T.Tensor((N,), dtype), B: T.Tensor(new_shape, new_dtype if new_dtype else dtype), )
testing/python/language/test_tilelang_language_view.py:22
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/language/test_tilelang_language_tma_copy.py:40
Functionmain
( A: T.Tensor((N,), dtype), B: T.Tensor((N,), dtype), )
testing/python/language/test_tilelang_language_clamp.py:15
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M, N), dtype), )
testing/python/language/test_tilelang_language_if_range.py:12
Functionmain
(flag: T.Buffer((1,), "int32"), out: T.Buffer((1,), "int32"))
testing/python/cpu/test_tilelang_cpu_while_repro.py:22
Functionmain
( A: T.Tensor(A_shape, dtype), B: T.Tensor(B_shape, dtype), C: T.Tensor((M, N), dtype)
testing/python/cpu/test_tilelang_cpu_tgemm.py:22
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/amd/test_tilelang_gemm_mfma_preshuffle.py:160
Functionmain
(Out: T.Tensor((64,), "int32"))
testing/python/amd/test_tilelang_hip_codegen.py:282
Functionmain
( src: T.Tensor((QN,), T.uint8), dst: T.Tensor((QN,), T.uint8), )
testing/python/amd/test_tilelang_mxfp4_gfx950.py:72
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/amd/test_tilelang_gemm_mfma_intrinsic.py:83
Functionmain
(A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N), out_dtype))
testing/python/amd/test_tilelang_test_amd.py:31
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/amd/test_tilelang_gfx950_copy_async.py:45
Functionmain
()
testing/python/issue/test_tilelang_issue_merge_if.py:10
Functionmain
(A: T.Tensor((16, 14), dtype=dtype), B: T.Tensor((16, 448), dtype=dtype))
testing/python/issue/test_tilelang_issue_1374.py:13
Functionmain
( pos: T.Tensor( [ S, ], pos_t
testing/python/issue/test_tilelang_issue_1115.py:15
Functionmain
(A: T.Tensor((256, 512), T.float4_e2m1fn))
testing/python/issue/test_tilelang_issue_2292.py:38
Functionmain
( A: T.Tensor[(2, 512), T.float32], B: T.Tensor[(2, 512), T.float32], C: T
testing/python/issue/test_tilelang_issue_1734.py:12
Functionmain
( A: T.Tensor((8,), T.float8_e4m3fn), )
testing/python/issue/test_tilelang_issue_1601.py:11
Functionmain
( Data: T.Tensor((M,), dtype), )
testing/python/issue/test_tilelang_issue_1549.py:15
Functionmain
(out: T.Tensor[(512,), dtype])
testing/python/issue/test_tilelang_issue_1604.py:12
Functionmain
( A: T.Tensor((M, K), dtype), B: T.Tensor((N, K), dtype), C: T.Tensor((M, N), dtype),
testing/python/issue/test_tilelang_issue_96.py:9
Functionmain
( A_sparse: T.Tensor((M, K // 2), T.float16), E: T.Tensor((M, K // e_factor), T.int8),
testing/python/issue/test_tilelang_issue_ws_simt_copy_full_producer_extent.py:31
Functionmain
( A: T.Tensor((M, K), T.float16), B: T.Tensor((K, N), T.float16), C: T.Tensor((M, N),
testing/python/issue/test_tilelang_issue_1697.py:9
Functionmain
(score: T.Tensor((16,), T.float32))
testing/python/issue/test_tilelang_issue_2307.py:9
Functionmain
( A: T.Tensor((length,), dtype), B: T.Tensor((length,), dtype), )
testing/python/analysis/test_tilelang_nested_loop_checker.py:34
Functionmain
( data: T.Tensor((128, A), dtype), # type: ignore )
testing/python/analysis/test_tilelang_fragment_loop_checker.py:12
Functionmain
( A: T.Tensor((M, N), dtype), B: T.Tensor((M, N), dtype), )
testing/python/fastmath/test_mathops_fastmath.py:69
Functionmain
( A: T.Tensor(A_shape, in_dtype), B: T.Tensor(B_shape, in_dtype), C: T.Tensor((M, N),
testing/python/jit/test_tilelang_jit_cutedsl.py:43
← previousnext →5,001–5,100 of 7,254, ranked by callers