Merge branch 'develop' into codegen_hiprtc

a11cf2c6 · arai713 · GitHub · a72e9efa · 64d5c4d6 · a11cf2c6
Unverified Commit a11cf2c6 authored Jan 24, 2025 by arai713 Committed by GitHub Jan 24, 2025
20 changed files
--- a/include/ck_tile/core/arch/arch.hpp
+++ b/include/ck_tile/core/arch/arch.hpp
--- a/include/ck_tile/core/numeric/bfloat16.hpp
+++ b/include/ck_tile/core/numeric/bfloat16.hpp
--- a/include/ck_tile/core/utility/amd_address_space.hpp
+++ b/include/ck_tile/core/utility/amd_address_space.hpp
--- a/include/ck_tile/core/utility/type_traits.hpp
+++ b/include/ck_tile/core/utility/type_traits.hpp
--- a/include/ck_tile/core/utility/unary_element_function.hpp
+++ b/include/ck_tile/core/utility/unary_element_function.hpp
--- a/include/ck_tile/host/check_err.hpp
+++ b/include/ck_tile/host/check_err.hpp
--- a/include/ck_tile/host/convolution_host_tensor_descriptor_helper.hpp
+++ b/include/ck_tile/host/convolution_host_tensor_descriptor_helper.hpp
--- a/include/ck_tile/host/host_tensor.hpp
+++ b/include/ck_tile/host/host_tensor.hpp
--- a/include/ck_tile/host/reference/reference_fused_moe.hpp
+++ b/include/ck_tile/host/reference/reference_fused_moe.hpp
--- a/include/ck_tile/host/reference/reference_rmsnorm2d_fwd.hpp
+++ b/include/ck_tile/host/reference/reference_rmsnorm2d_fwd.hpp
--- a/include/ck_tile/host/reference/reference_rowwise_quantization2d.hpp
+++ b/include/ck_tile/host/reference/reference_rowwise_quantization2d.hpp
--- a/include/ck_tile/ops/elementwise/unary_element_wise_operation.hpp
+++ b/include/ck_tile/ops/elementwise/unary_element_wise_operation.hpp
--- a/include/ck_tile/ops/epilogue/dynamic_quant_epilogue.hpp
+++ b/include/ck_tile/ops/epilogue/dynamic_quant_epilogue.hpp
--- a/include/ck_tile/ops/flatmm/block/flatmm_32x512x128_1x4x1_16x16x32.hpp
+++ b/include/ck_tile/ops/flatmm/block/flatmm_32x512x128_1x4x1_16x16x32.hpp
--- a/include/ck_tile/ops/flatmm/block/flatmm_sn_32x128x512_1x4x1_16x16x32.hpp
+++ b/include/ck_tile/ops/flatmm/block/flatmm_sn_32x128x512_1x4x1_16x16x32.hpp
--- a/include/ck_tile/ops/flatmm/block/flatmm_sn_32x128x512_1x4x1_16x16x32_itl.hpp
+++ b/include/ck_tile/ops/flatmm/block/flatmm_sn_32x128x512_1x4x1_16x16x32_itl.hpp
--- a/include/ck_tile/ops/flatmm/block/uk/flatmm_sn_uk_gfx9_32x128x512_1x4x1_16x16x16.inc
+++ b/include/ck_tile/ops/flatmm/block/uk/flatmm_sn_uk_gfx9_32x128x512_1x4x1_16x16x16.inc
--- a/include/ck_tile/ops/flatmm/block/uk/flatmm_sn_uk_gfx9_32x128x512_1x4x1_16x16x16_itl.inc
+++ b/include/ck_tile/ops/flatmm/block/uk/flatmm_sn_uk_gfx9_32x128x512_1x4x1_16x16x16_itl.inc
--- a/include/ck_tile/ops/flatmm/block/uk/flatmm_uk_gfx9_32x512x128_1x1x1_16x16x16.inc
+++ b/include/ck_tile/ops/flatmm/block/uk/flatmm_uk_gfx9_32x512x128_1x1x1_16x16x16.inc
--- a/include/ck_tile/ops/fused_moe/kernel/fused_moegemm_kernel.hpp
+++ b/include/ck_tile/ops/fused_moe/kernel/fused_moegemm_kernel.hpp