Sami Remes 919096fde8 [rocm-libraries] ROCm/rocm-libraries#7935 (commit 5c96097) ...

[CK] Allow skipping split-K C-buffer zero-init in
 xdl_cshuffle blockscale GEMM (#7935)

Add a `skip_zero_init` flag (default false) to the Problem/Argument of
the xdl_cshuffle block-scale GEMM device ops (multiple_d ab_scale and
blockscale b-preshuffle). When the flag is set, the device invoker skips
the internal hipMemsetAsync that zeroes p_c_grid before the KBatch > 1
split-K atomic-accumulation path. The flag is declared on the gridwise
Problem struct (inherited by Argument), so it is visible on both the
rotating-cache (arg_) and the normal (arg) launch paths in each device
op.

Why: callers that already pre-zero the output buffer otherwise pay for a
redundant device-wide memset before split-K atomic accumulation. Gating
the memset behind an opt-in flag lets such callers avoid the duplicate
work. Because the flag defaults to false, every existing call site is
unaffected and the observable behavior is unchanged.

## Motivation

<!-- Explain the purpose of this PR and the goals it aims to achieve.
-->

## Technical Details

<!-- Explain the changes along with any relevant GitHub links. -->

## Test Plan

<!-- Explain any relevant testing done to verify this PR. -->

## Test Result

<!-- Briefly summarize test outcomes. -->

## Submission Checklist

- [ ] Look over the contributing guidelines at
https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

Co-authored-by: Cursor <cursoragent@cursor.com>

2026-06-02 13:08:46 +00:00

..

codegen_device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_avgpool2d_bwd_nhwc_nhwc.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_avgpool3d_bwd_ndhwc_ndhwc.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_batched_contraction_multiple_d_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_contraction_multiple_d_wmma_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_batched_contraction_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_e_permute_xdl.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_batched_gemm_gemm_wmma_cshuffle_v3_common.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_gemm_wmma_cshuffle_v3.hpp

Remove code duplications in batched gemm (multi D) gemm (multi D) wmma (#3617)

2026-01-26 10:20:30 -08:00

device_batched_gemm_gemm_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_multi_d_xdl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_multiple_d_dl.hpp

Add support for RDNA1 GPUs (#3220)

2025-11-20 10:45:57 -08:00

device_batched_gemm_multiple_d_gemm_multiple_d_wmma_cshuffle_v3.hpp

Remove code duplications in batched gemm (multi D) gemm (multi D) wmma (#3617)

2026-01-26 10:20:30 -08:00

device_batched_gemm_multiple_d_gemm_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_batched_gemm_multiple_d_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_batched_gemm_multiple_d_xdl_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_batched_gemm_reduce_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_reduce_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_softmax_gemm_permute_wmma_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_softmax_gemm_permute_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_softmax_gemm_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batched_gemm_wmma_cshuffle_v3_b_scale.hpp

Remove code duplications in batched gemm wmma (#3580)

2026-01-23 12:39:03 -08:00

device_batched_gemm_wmma_cshuffle_v3_common.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_batched_gemm_wmma_cshuffle_v3.hpp

Remove code duplications in batched gemm wmma (#3580)

2026-01-23 12:39:03 -08:00

device_batched_gemm_xdl_fpAintB_b_scale.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_batched_gemm_xdl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batchnorm_backward_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batchnorm_forward_impl_obsolete.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_batchnorm_forward_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_cgemm_4gemm_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_column_to_image_impl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_contraction_multiple_abd_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_contraction_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6022 (commit 54b284a)

2026-03-31 08:18:11 -07:00

device_contraction_utils.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_conv2d_backward_weight_xdl_c_shuffle_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_conv2d_bwd_data_xdl_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_conv2d_fwd_xdl_c_shuffle_bias_activation_add_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_conv2d_fwd_xdl_c_shuffle_bias_activation_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_conv2d_fwd_xdl_c_shuffle_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_conv2d_fwd_xdl_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_conv3d_fwd_naive_ndhwc_kzyxc_ndhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_conv3d_fwd_xdl_ndhwc_kzyxc_ndhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_convnd_bwd_data_nwc_kxc_nwk_dl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_convnd_bwd_data_nwc_kxc_nwk_xdl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_elementwise_dynamic_vector_dims_impl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_elementwise_normalization_impl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_elementwise_scale_impl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_fpAintB_gemm_wmma.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_bias_add_reduce_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_bias_add_reduce_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_dl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_gemm_dpp.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_abd_wmma_cshuffle_v3.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_gemm_multiple_abd_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_d_dl.hpp

Add support for RDNA1 GPUs (#3220)

2025-11-20 10:45:57 -08:00

device_gemm_multiple_d_layernorm_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_multiple_d_layernorm_xdl_cshuffle.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_gemm_multiple_d_multiple_r_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_multiple_d_wmma_cshuffle_v3_ab_scale.hpp

Wmma support for gemm_ab_scale (#3314)

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

device_gemm_multiple_d_wmma_cshuffle_v3_b_preshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_d_wmma_cshuffle_v3_blockscale_bpreshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_d_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_multiple_d_wmma_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_d_xdl_cshuffle_lds_direct_load.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_d_xdl_cshuffle_v3_ab_scale.hpp

[rocm-libraries] ROCm/rocm-libraries#7935 (commit 5c96097)

2026-06-02 13:08:46 +00:00

device_gemm_multiple_d_xdl_cshuffle_v3_b_preshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_multiple_d_xdl_cshuffle_v3_blockscale_bpreshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7935 (commit 5c96097)

2026-06-02 13:08:46 +00:00

device_gemm_multiple_d_xdl_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#6761 (commit d19f6f1)

2026-05-27 18:55:15 +00:00

device_gemm_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_reduce_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_reduce_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_wmma_cshuffle_v3_b_preshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_wmma_cshuffle_v3_b_scale.hpp

Wmma support for gemm_ab_scale (#3314)

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

device_gemm_wmma_cshuffle_v3_common.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_gemm_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_wmma_cshuffle_v3r1.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_wmma.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_lds_direct_load.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_streamk_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_v2.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_v3_b_preshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_v3_b_scale.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_v3_mx.hpp

[rocm-libraries] ROCm/rocm-libraries#6761 (commit d19f6f1)

2026-05-27 18:55:15 +00:00

device_gemm_xdl_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle_v3r1.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_layernorm_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_xdl_skip_b_lds.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_xdl_splitk_c_shuffle_lds_direct_load.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_splitk_c_shuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_streamk.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_gemm_xdl_waveletmodel_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_gemm_xdl.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_contraction_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_conv_bwd_data_multiple_d_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_data_multiple_d_wmma_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_data_multiple_d_xdl_cshuffle_v1.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_data_multiple_d_xdl_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7847 (commit b995ef2)

2026-05-29 14:00:06 +00:00

device_grouped_conv_bwd_weight_dl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_grouped_conv_bwd_weight_explicit.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_weight_multiple_d_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_grouped_conv_bwd_weight_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_conv_bwd_weight_two_stage_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_weight_two_stage_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7847 (commit b995ef2)

2026-05-29 14:00:06 +00:00

device_grouped_conv_bwd_weight_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_weight_wmma_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_weight_xdl_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7847 (commit b995ef2)

2026-05-29 14:00:06 +00:00

device_grouped_conv_bwd_weight_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_bwd_weight_xdl_waveletmodel_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_fwd_dl_nhwc_kyxc_nhwk.hpp

Add support for RDNA1 GPUs (#3220)

2025-11-20 10:45:57 -08:00

device_grouped_conv_fwd_multiple_abd_wmma_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_grouped_conv_fwd_multiple_abd_xdl_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#7847 (commit b995ef2)

2026-05-29 14:00:06 +00:00

device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_grouped_conv_fwd_multiple_d_multiple_r_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_grouped_conv_fwd_multiple_d_multiple_r.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_grouped_conv_fwd_multiple_d_wmma_cshuffle_v3_large_tensor.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00

device_grouped_conv_fwd_multiple_d_wmma_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_grouped_conv_fwd_multiple_d_xdl_large_tensor_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_conv_utils.hpp

[rocm-libraries] ROCm/rocm-libraries#7847 (commit b995ef2)

2026-05-29 14:00:06 +00:00

device_grouped_gemm_fixed_nk_common.hpp

[rocm-libraries] ROCm/rocm-libraries#4961 (commit 6c3969a)

2026-04-20 12:24:59 +00:00

device_grouped_gemm_multi_abd_wmma_fixed_nk.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_grouped_gemm_multi_abd_xdl_fixed_nk.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_gemm_multiple_d_dl.hpp

Add support for RDNA1 GPUs (#3220)

2025-11-20 10:45:57 -08:00

device_grouped_gemm_multiple_d_splitk_xdl_cshuffle_two_stage.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_gemm_multiple_d_wmma_cshuffle_tile_loop_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_grouped_gemm_multiple_d_xdl_cshuffle_tile_loop.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_gemm_softmax_gemm_permute_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_gemm_wmma_fixed_nk.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_grouped_gemm_wmma_splitk_cshuffle_v3.hpp

[rocm-libraries] ROCm/rocm-libraries#4871 (commit 7d4c040)

2026-05-22 18:39:01 +00:00

device_grouped_gemm_xdl_fixed_nk.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_gemm_xdl_splitk_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_gemm_xdl.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_grouped_query_attention_forward_wmma.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_image_to_column_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_max_pool_bwd_impl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_moe_gemm_blockscale.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_moe_gemm.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_moe_mx_gemm_bns.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_moe_mx_gemm_bpreshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_moe_mx_gemm.hpp

[rocm-libraries] ROCm/rocm-libraries#6978 (commit e58096d)

2026-05-15 06:46:51 -07:00

device_multi_query_attention_forward_wmma.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_multiple_reduce_multiblock.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_multiple_reduce_threadwise.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_normalization_bwd_data_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_normalization_bwd_gamma_beta_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_normalization_fwd_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_normalization_fwd_splitk_impl.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_permute_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_pool2d_fwd_nhwc_nhwc.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_pool3d_fwd_ndhwc_ndhwc.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_put_element_impl.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_reduce_common.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_reduce_multiblock.hpp

chore(copyright): update copyright header for include directory (#3224)

2025-11-18 10:17:18 -08:00

device_reduce_threadwise_multi_d.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_reduce_threadwise.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_softmax_impl.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

device_sparse_embeddings_forward_layernorm.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

device_splitk_contraction_multiple_d_xdl_cshuffle.hpp

[rocm-libraries] ROCm/rocm-libraries#7760 (commit a61bc76)

2026-05-27 06:56:58 -07:00

split_k_arg.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

split_k_offset_utils.hpp

[CK] Allow tensors larger than 2GB in grouped conv bwd weight (#3169)

2026-01-08 08:02:02 +01:00

split_k_utils.hpp

[rocm-libraries] ROCm/rocm-libraries#7631 (commit d591a7c)

2026-05-27 08:21:54 +00:00