mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 10:09:41 +00:00

Files

Yi DING fb72fea980 [CK_TILE] Add FP8xF4 Flatmm (#3401 )

* Refactor policy

* fix a bank conflict

* Enable mixed mx flatmm

* Update

[ROCm/composable_kernel commit: 57e1e4a848]

2025-12-17 10:01:48 +08:00

01_gemm

…

02_gemm_bilinear

…

03_gemm_bias_relu

…

04_gemm_add_add_fastgelu

…

09_convnd_fwd

…

10_convnd_fwd_multiple_d_multiple_reduce

…

11_convnd_fwd_bias

…

12_reduce

…

13_pool2d_fwd

…

14_gemm_quantization

…

15_grouped_gemm

…

16_gemm_multi_d_multi_reduces

…

17_convnd_bwd_data

…

18_batched_gemm_reduce

…

19_binary_elementwise

…

20_grouped_conv_bwd_weight

…

21_gemm_layernorm

…

22_cgemm

…

23_softmax

…

24_batched_gemm

…

25_gemm_bias_e_permute

…

26_contraction

[ck][gfx12] support contraction on gfx12 (#3421 )

2025-12-15 07:16:01 -08:00

27_layernorm2d_fwd

…

28_grouped_gemm_bias_e_permute

…

29_batched_gemm_bias_e_permute

…

30_grouped_conv_fwd_multiple_d

…

31_batched_gemm_gemm

…

32_batched_gemm_scale_softmax_gemm

…

33_multiple_reduce

…

34_batchnorm

…

35_splitK_gemm

…

36_sparse_embedding

…

37_batched_gemm_add_add_relu_gemm_add

…

38_grouped_conv_bwd_data_multiple_d

…

39_permute

…

40_conv2d_fwd_quantization

…

41_grouped_conv_conv_fwd

…

42_groupnorm_fwd

…

43_splitk_gemm_bias_e_permute

…

44_elementwise_permute

…

45_elementwise_normalization

…

46_gemm_add_multiply

…

47_gemm_bias_softmax_gemm_permute

…

48_pool3d_fwd

…

49_maxpool2d_bwd

…

50_put_element

…

51_avgpool3d_bwd

…

52_im2col_col2im

…

53_layernorm2d_bwd

…

54_groupnorm_bwd

…

59_grouped_gemm_multi_ABD

…

60_gemm_multi_ABD

…

61_contraction_multi_ABD

…

62_convnd_activ

…

63_layernorm4d_fwd

…

64_fpAintB_gemm

…

65_gemm_multiply_multiply

Wmma support for gemm_ab_scale (#3314 )

2025-12-11 09:06:20 +01:00

66_complex_contraction_bilinear

…

67_gemm_microscaling

…

68_gemm_add

…

69_gemm_add_relu

…

ck_tile

[CK_TILE] Add FP8xF4 Flatmm (#3401 )

2025-12-17 10:01:48 +08:00

CMakeLists.txt

…

README.md

…

test_old_ck_gpu_reference.cpp

chore: update copyright header for misc files (#3402 )

2025-12-11 08:25:29 -08:00

README.md

Back to the main page

Composable Kernel examples