Enrico Degregori 12225ce645 Wmma support for multiple ABD GEMM (#2803) ...

* multi_abd wmma support:

 - Add multiple A and B support to multiple D implementation (gridwise level)
 - Add multi_abd GEMM (device level)
 - Add instances (xdl parity)
 - Add tests (both xdl and wmma)
 - Add examples
 - Add ckProfiler support (both xdl and wmma)

* Fix bug in device print function

* Fix unused template parameter

* Fix batched gemm for multiABD gridwise implementation

* Fix gemm_universal_reduce with multiABDs gridwise implementation

---------

Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>

[ROCm/composable_kernel commit: 3d29bff2f0]

2025-09-22 18:49:06 -07:00

..

CMakeLists.txt

Wmma support for multiple ABD GEMM (#2803)

2025-09-22 18:49:06 -07:00

profile_avg_pool2d_bwd.cpp

Pool2d max/avg kernel in the BWD version (#1494)

2024-09-12 11:47:52 +02:00

profile_avg_pool3d_bwd.cpp

MaxPool & AvgPool bwd instances, test, ckProfiler, client example (#861)

2023-08-31 21:01:50 +08:00

profile_batched_gemm_add_relu_gemm_add.cpp

…

profile_batched_gemm_b_scale.cpp

Fix build errors on windows (#2456)

2025-07-16 07:58:23 -07:00

profile_batched_gemm_gemm.cpp

…

profile_batched_gemm_multi_d.cpp

Fix argument order for calls to profile_batched_gemm_impl() (#2277)

2025-06-17 19:29:09 -07:00

profile_batched_gemm_reduce.cpp

…

profile_batched_gemm.cpp

Fix argument order for calls to profile_batched_gemm_impl() (#2277)

2025-06-17 19:29:09 -07:00

profile_batchnorm_bwd.cpp

…

profile_batchnorm_fwd.cpp

Batchnorm splitk single kernel (#771)

2023-07-06 10:58:55 -05:00

profile_batchnorm_infer.cpp

…

profile_contraction_bilinear.cpp

upgrade from clang-format-12 to clang-format-18 (#2568)

2025-07-28 11:34:07 -07:00

profile_contraction_scale.cpp

upgrade from clang-format-12 to clang-format-18 (#2568)

2025-07-28 11:34:07 -07:00

profile_conv_bwd_data.cpp

[HotFix] add config and version files to pass on build info (#856)

2023-08-23 11:36:17 -07:00

profile_conv_fwd_bias_relu_add.cpp

…

profile_conv_fwd_bias_relu.cpp

…

profile_conv_fwd.cpp

…

profile_conv_tensor_rearrange.cpp

Add support for groups in Img2Col/Col2Img (#1007)

2023-10-31 10:46:32 +01:00

profile_gemm_ab_scale.cpp

Add other layouts for FP8 block scaled gemm (#2665)

2025-08-18 01:46:10 -07:00

profile_gemm_add_add_fastgelu.cpp

…

profile_gemm_add_fastgelu.cpp

Add support for mixed-precision f16bf16_int8 gemm (#1127)

2024-02-07 15:54:13 +01:00

profile_gemm_add_multiply.cpp

Wmma support for multiple Ds based GEMMs (#2613)

2025-09-05 16:31:08 +02:00

profile_gemm_add_relu_add_layernorm.cpp

…

profile_gemm_add_relu.cpp

Add support for mixed-precision f16bf16_int8 gemm (#1127)

2024-02-07 15:54:13 +01:00

profile_gemm_add_silu.cpp

Wmma support for multiple Ds based GEMMs (#2613)

2025-09-05 16:31:08 +02:00

profile_gemm_add.cpp

Add support for mixed-precision f16bf16_int8 gemm (#1127)

2024-02-07 15:54:13 +01:00

profile_gemm_b_scale.cpp

Fix build errors on windows (#2456)

2025-07-16 07:58:23 -07:00

profile_gemm_bias_add_reduce.cpp

…

profile_gemm_bilinear.cpp

[Navi3x] Add fp16/int8 wmma conv forward instances (#746)

2023-09-07 21:59:26 -05:00

profile_gemm_blockscale_wp.cpp

Add MoE & FP8 Blockscale WP Kernels for GFX950 (#2297)

2025-06-12 09:25:59 +08:00

profile_gemm_fastgelu.cpp

…

profile_gemm_multi_abd.cpp

Wmma support for multiple ABD GEMM (#2803)

2025-09-22 18:49:06 -07:00

profile_gemm_multiply_add.cpp

Refactor f8_t, add bf8_t (#792)

2023-09-12 17:04:27 -05:00

profile_gemm_multiply_multiply_wp.cpp

[Block Scale GEMM] Optimized block scale gemm (#1950)

2025-03-11 10:11:21 -07:00

profile_gemm_multiply_multiply.cpp

Extend XDL kernel to Support RDNA3/4 - Part 5 (#2725)

2025-09-15 10:59:25 -07:00

profile_gemm_mx.cpp

Optimized GEMMs for MX FP4/8 (#2294)

2025-06-05 13:54:15 -06:00

profile_gemm_quantization.cpp

Added wmma support for gemm quantization: (#2841)

2025-09-16 16:23:29 -07:00

profile_gemm_reduce.cpp

…

profile_gemm_splitk.cpp

Add an option to change the number of warm-up cycles and iterations. (#1124)

2024-01-09 09:43:08 -08:00

profile_gemm_streamk.cpp

initial stream-k implementation with example (#699)

2023-07-26 14:18:15 -05:00

profile_gemm_universal_batched.cpp

Add SplitK support into Batched GEMM V3 (#1729)

2024-12-13 21:08:35 +01:00

profile_gemm_universal_preshuffle.cpp

Extend XDL kernel to Support RDNA3/4 - Part 5 (#2725)

2025-09-15 10:59:25 -07:00

profile_gemm_universal_reduce.cpp

Universal gemm splitk using reduce (with multi-d) (#1341)

2024-07-19 22:01:22 +08:00

profile_gemm_universal_streamk.cpp

Wmma support for multiple Ds based GEMMs (#2613)

2025-09-05 16:31:08 +02:00

profile_gemm_universal.cpp

Extend XDL kernel to Support RDNA3/4 - Part 5 (#2725)

2025-09-15 10:59:25 -07:00

profile_gemm.cpp

Fixing most of the cppcheck errors. (#1142)

2024-01-24 13:47:48 -08:00

profile_grouped_conv_bwd_data.cpp

Integrate universal gemm with conv bwd data and add SplitK (#1315)

2025-04-28 23:54:49 +02:00

profile_grouped_conv_bwd_weight.cpp

Automatic deduction of split-K value for grouped convolution (#2491)

2025-07-31 12:08:45 +02:00

profile_grouped_conv_fwd_bias_clamp.cpp

Add v3 support for Groupd fwd conv+bias+clamp & ckProfiler (#2463)

2025-07-25 10:34:31 +02:00

profile_grouped_conv_fwd_clamp.cpp

Add v3 support for Groupd fwd conv+bias+clamp & ckProfiler (#2463)

2025-07-25 10:34:31 +02:00

profile_grouped_conv_fwd_outelementop.cpp

Add ckProfiler support for forward 3D convolutions with OUT element-wise operations. (#1354)

2024-07-08 10:55:54 -07:00

profile_grouped_conv_fwd.cpp

TF32 POC in Conv3d on MI30x platform #2763 (second attempt) (#2852)

2025-09-17 14:50:15 -07:00

profile_grouped_gemm_fastgelu.cpp

…

profile_grouped_gemm_fixed_nk.cpp

Added bf16 instances grouped gemm fixed nk (#1825)

2025-01-20 09:13:09 -08:00

profile_grouped_gemm_multiply_tile_loop.cpp

Switch to universal gemm in grouped gemm tile loop (#1335)

2024-06-18 09:01:49 -05:00

profile_grouped_gemm_tile_loop.cpp

Grouped GEMM Multiple D tile loop. (#1247)

2024-04-25 15:12:53 -05:00

profile_grouped_gemm.cpp

Polished Grouped GEMM APIs and new BF16 instances (#1600)

2024-11-27 13:02:44 +01:00

profile_groupnorm_bwd_data.cpp

layernorm and groupnorm backward data (#1083)

2023-12-19 04:23:11 +08:00

profile_groupnorm_bwd_gamma_beta.cpp

layernorm & groupnorm bwd gamma beta (#1133)

2024-01-25 19:53:15 +08:00

profile_groupnorm_fwd.cpp

layernorm and groupnorm backward data (#1083)

2023-12-19 04:23:11 +08:00

profile_layernorm_bwd_data.cpp

layernorm and groupnorm backward data (#1083)

2023-12-19 04:23:11 +08:00

profile_layernorm_bwd_gamma_beta.cpp

layernorm & groupnorm bwd gamma beta (#1133)

2024-01-25 19:53:15 +08:00

profile_layernorm_fwd.cpp

Fix F16 type (#1583)

2024-11-06 11:32:44 -08:00

profile_max_pool2d_bwd.cpp

Pool2d max/avg kernel in the BWD version (#1494)

2024-09-12 11:47:52 +02:00

profile_max_pool2d_fwd.cpp

Add pool2d int8 and fp8 instances (#1508)

2024-09-13 10:18:21 -07:00

profile_max_pool3d_bwd.cpp

MaxPool & AvgPool bwd instances, test, ckProfiler, client example (#861)

2023-08-31 21:01:50 +08:00

profile_permute_scale.cpp

Add elementwise with dynamic vector dim (#1198)

2024-03-22 10:40:43 +01:00

profile_pool3d_fwd.cpp

Extend pool3d fwd avg, max operations by f8_t, int8_t types (#1483)

2024-09-17 15:57:10 +02:00

profile_reduce.cpp

…

profile_softmax.cpp

Padded Generic Kernel Instance (#730)

2023-06-16 23:43:11 -05:00

profile_transpose.cpp

Refactor elementwise kernels (#1222)

2024-04-19 13:31:17 +02:00

profiler_operation_registry.hpp

…

profiler.cpp

…