ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-19 14:29:05 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
352dee5225cede21e82bb96f530425e54139f251
composable_kernel/example
History
Emily Martins 352dee5225 Fix CK Tile Stream-K BF16 Validation Errors (#3039) 
Prior to this change, the number of accumulations passed into
calculate_rtol_atol was 1. That said, in most cases, this is not correct
when there are multiple workgroups contributing to the same macro tile
in C.

This change ensures uses the function estimate_num_wgs_per_tile, which
was extracted into a common file and generalized, to estimate the number
of workgroups per macro tile. This estimate is passed into
calculate_rtol_atol to ensure we get a better relative and absolute
tolerance.
2025-10-17 09:33:38 -07:00
..
01_gemm
Wave Tile Transfer supporting global load with transpose (#3027)
2025-10-16 11:33:56 -07:00
02_gemm_bilinear
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
03_gemm_bias_relu
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
04_gemm_add_add_fastgelu
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
09_convnd_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
10_convnd_fwd_multiple_d_multiple_reduce
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
11_convnd_fwd_bias
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
12_reduce
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
13_pool2d_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
14_gemm_quantization
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
15_grouped_gemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
16_gemm_multi_d_multi_reduces
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
17_convnd_bwd_data
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
18_batched_gemm_reduce
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
19_binary_elementwise
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
20_grouped_conv_bwd_weight
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
21_gemm_layernorm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
22_cgemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
23_softmax
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
24_batched_gemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
25_gemm_bias_e_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
26_contraction
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
27_layernorm2d_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
28_grouped_gemm_bias_e_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
29_batched_gemm_bias_e_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
30_grouped_conv_fwd_multiple_d
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
31_batched_gemm_gemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
32_batched_gemm_scale_softmax_gemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
33_multiple_reduce
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
34_batchnorm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
35_splitK_gemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
36_sparse_embedding
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
37_batched_gemm_add_add_relu_gemm_add
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
38_grouped_conv_bwd_data_multiple_d
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
39_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
40_conv2d_fwd_quantization
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
41_grouped_conv_conv_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
42_groupnorm_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
43_splitk_gemm_bias_e_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
44_elementwise_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
45_elementwise_normalization
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
46_gemm_add_multiply
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
47_gemm_bias_softmax_gemm_permute
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
48_pool3d_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
49_maxpool2d_bwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
50_put_element
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
51_avgpool3d_bwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
52_im2col_col2im
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
53_layernorm2d_bwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
54_groupnorm_bwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
59_grouped_gemm_multi_ABD
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
60_gemm_multi_ABD
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
61_contraction_multi_ABD
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
62_convnd_activ
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
63_layernorm4d_fwd
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
64_fpAintB_gemm
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
65_gemm_multiply_multiply
implement device batched gemm b scale for wmma (#2825)
2025-10-16 11:00:42 -07:00
66_complex_contraction_bilinear
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
67_gemm_microscaling
[DOCS] Documentation Addition (Readme updates) (#2495)
2025-10-16 03:10:57 -07:00
68_gemm_add
Extend XDL kernel to Support RDNA3/4 - Part 4 (#2724)
2025-09-12 08:17:07 -07:00
69_gemm_add_relu
Extend XDL kernel to Support RDNA3/4 - Part 4 (#2724)
2025-09-12 08:17:07 -07:00
ck_tile
Fix CK Tile Stream-K BF16 Validation Errors (#3039)
2025-10-17 09:33:38 -07:00
CMakeLists.txt
Extend XDL kernel to Support RDNA3/4 - Part 5 (#2725)
2025-09-15 10:59:25 -07:00
README.md
Add basic documentation structure (#1715)
2024-12-04 00:46:47 +01:00

README.md

Back to the main page

Composable Kernel examples

Reference in New Issue View Git Blame Copy Permalink