Skip to content

GitLab

Commit 1fb6a1e0 authored by Adam Osewski's avatar Adam Osewski
Browse files

Don't pass kbatch to CalculateKPadded.

parent a3e2ddba
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
include/ck/tensor_operation/gpu/device/impl/device_grouped_gemm_xdl_splitk_cshuffle.hpp
View file @ 1fb6a1e0
...@@ -262,7 +262,7 @@ struct DeviceGroupedGemmXdlSplitKCShuffle : public DeviceGroupedGemmSplitK<ALayo ...@@ -262,7 +262,7 @@ struct DeviceGroupedGemmXdlSplitKCShuffle : public DeviceGroupedGemmSplitK<ALayo
const index_t m_padded = GridwiseGemm::CalculateMPadded(M); const index_t m_padded = GridwiseGemm::CalculateMPadded(M);
const index_t n_padded = GridwiseGemm::CalculateNPadded(N); const index_t n_padded = GridwiseGemm::CalculateNPadded(N);
const index_t k_padded = GridwiseGemm::CalculateKPadded(K, K_BATCH); const index_t k_padded = GridwiseGemm::CalculateKPadded(K);
const index_t k0 = GridwiseGemm::CalculateK0(K, K_BATCH); const index_t k0 = GridwiseGemm::CalculateK0(K, K_BATCH);
const auto c_grid_desc_m_n = const auto c_grid_desc_m_n =
...@@ -316,7 +316,7 @@ struct DeviceGroupedGemmXdlSplitKCShuffle : public DeviceGroupedGemmSplitK<ALayo ...@@ -316,7 +316,7 @@ struct DeviceGroupedGemmXdlSplitKCShuffle : public DeviceGroupedGemmSplitK<ALayo
auto& karg = gemm_kernel_args_[i].karg_; auto& karg = gemm_kernel_args_[i].karg_;
const index_t k_padded = GridwiseGemm::CalculateKPadded(karg.K, K_BATCH); const index_t k_padded = GridwiseGemm::CalculateKPadded(karg.K);
const index_t k0 = GridwiseGemm::CalculateK0(karg.K, K_BATCH); const index_t k0 = GridwiseGemm::CalculateK0(karg.K, K_BATCH);
const auto c_grid_desc_m_n = GridwiseGemm::MakeCGridDescriptor_M_N( const auto c_grid_desc_m_n = GridwiseGemm::MakeCGridDescriptor_M_N(
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment