From d0125f7c26565bbe3606398e5a1b212489f41edd Mon Sep 17 00:00:00 2001 From: Evgeny Mankov Date: Wed, 18 Dec 2024 17:41:07 +0100 Subject: [PATCH] [HIPIFY][Tensor] Minor fixes and formatting --- bin/hipify-perl | 2 +- src/CUDA2HIP_TENSOR_API_functions.cpp | 199 ++++---- src/CUDA2HIP_TENSOR_API_types.cpp | 667 +++++++++++++------------- 3 files changed, 433 insertions(+), 435 deletions(-) diff --git a/bin/hipify-perl b/bin/hipify-perl index 06181010..36006f3f 100755 --- a/bin/hipify-perl +++ b/bin/hipify-perl @@ -7384,6 +7384,7 @@ sub simpleSubstitutions { subst("cusparseSparseToDenseAlg_t", "hipsparseSparseToDenseAlg_t", "type"); subst("cusparseStatus_t", "hipsparseStatus_t", "type"); subst("cutensorAlgo_t", "hiptensorAlgo_t", "type"); + subst("cutensorComputeType_t", "hiptensorComputeType_t", "type"); subst("cutensorContractionPlan_t", "hiptensorContractionPlan_t", "type"); subst("cutensorDataType_t", "hiptensorComputeType_t", "type"); subst("cutensorHandle_t", "hiptensorHandle_t", "type"); @@ -8694,7 +8695,6 @@ sub simpleSubstitutions { subst("cudaSuccess", "hipSuccess", "numeric_literal"); subst("cudaUserObjectNoDestructorSync", "hipUserObjectNoDestructorSync", "numeric_literal"); subst("cusolver_int_t", "int", "numeric_literal"); - subst("cutensorComputeType_t", "hiptensorComputeType_t", "numeric_literal"); subst("CUB_MAX", "CUB_MAX", "define"); subst("CUB_MIN", "CUB_MIN", "define"); subst("CUB_NAMESPACE_BEGIN", "BEGIN_HIPCUB_NAMESPACE", "define"); diff --git a/src/CUDA2HIP_TENSOR_API_functions.cpp b/src/CUDA2HIP_TENSOR_API_functions.cpp index 4bb5cc85..a6440012 100644 --- a/src/CUDA2HIP_TENSOR_API_functions.cpp +++ b/src/CUDA2HIP_TENSOR_API_functions.cpp @@ -1,5 +1,5 @@ /* -Copyright (c) 2015 - present Advanced Micro Devices, Inc. All rights reserved. +Copyright (c) 2024 - present Advanced Micro Devices, Inc. All rights reserved. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal @@ -23,111 +23,110 @@ THE SOFTWARE. #include "CUDA2HIP.h" const std::map CUDA_TENSOR_FUNCTION_MAP { - {"cutensorCreate", {"hiptensorCreate", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorDestroy", {"hiptensorDestroy", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorHandleResizePlanCache", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorHandleWritePlanCacheToFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorHandleReadPlanCacheFromFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorWriteKernelCacheToFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorReadKernelCacheFromFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreateTensorDescriptor", {"hiptensorInitTensorDescriptor", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorInitTensorDescriptor", {"hiptensorInitTensorDescriptor", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorDestroyTensorDescriptor", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreateElementwiseTrinary", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorElementwiseTrinaryExecute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreateElementwiseBinary", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorElementwiseBinaryExecute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreatePermutation", {"hiptensorPermutation", "", CONV_LIB_FUNC, API_TENSOR, 2, HIP_UNSUPPORTED}}, - {"cutensorPermutation", {"hiptensorPermutation", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorPermute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreateContraction", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorContraction", {"hiptensorContraction", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorDestroyOperationDescriptor", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorOperationDescriptorSetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorOperationDescriptorGetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreatePlanPreference", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorDestroyPlanPreference", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorPlanPreferenceSetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorPlanGetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorEstimateWorkspaceSize", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorCreatePlan", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorDestroyPlan", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorContract", {"hiptensorContraction", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorReduction", {"hiptensorReduction", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorCreateReduction", {"hiptensorReduction", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorReduce", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorGetErrorString", {"hiptensorGetErrorString", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorGetVersion", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, - {"cutensorGetCudartVersion", {"hiptensorGetHiprtVersion", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorLoggerSetCallback", {"hiptensorLoggerSetCallback", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorLoggerSetFile", {"hiptensorLoggerSetFile", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorLoggerOpenFile", {"hiptensorLoggerOpenFile", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorLoggerSetLevel", {"hiptensorLoggerSetLevel", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorLoggerSetMask", {"hiptensorLoggerSetMask", "", CONV_LIB_FUNC, API_TENSOR, 2}}, - {"cutensorLoggerForceDisable", {"hiptensorLoggerForceDisable", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorCreate", {"hiptensorCreate", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorDestroy", {"hiptensorDestroy", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorHandleResizePlanCache", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorHandleWritePlanCacheToFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorHandleReadPlanCacheFromFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorWriteKernelCacheToFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorReadKernelCacheFromFile", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreateTensorDescriptor", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorInitTensorDescriptor", {"hiptensorInitTensorDescriptor", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorDestroyTensorDescriptor", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreateElementwiseTrinary", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorElementwiseTrinaryExecute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreateElementwiseBinary", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorElementwiseBinaryExecute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreatePermutation", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorPermutation", {"hiptensorPermutation", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorPermute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreateContraction", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorContraction", {"hiptensorContraction", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorDestroyOperationDescriptor", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorOperationDescriptorSetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorOperationDescriptorGetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreatePlanPreference", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorDestroyPlanPreference", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorPlanPreferenceSetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorPlanGetAttribute", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorEstimateWorkspaceSize", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorCreatePlan", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorDestroyPlan", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorContract", {"hiptensorContraction", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorReduction", {"hiptensorReduction", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorCreateReduction", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorReduce", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorGetErrorString", {"hiptensorGetErrorString", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorGetVersion", {"", "", CONV_LIB_FUNC, API_TENSOR, 2, UNSUPPORTED}}, + {"cutensorGetCudartVersion", {"hiptensorGetHiprtVersion", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorLoggerSetCallback", {"hiptensorLoggerSetCallback", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorLoggerSetFile", {"hiptensorLoggerSetFile", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorLoggerOpenFile", {"hiptensorLoggerOpenFile", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorLoggerSetLevel", {"hiptensorLoggerSetLevel", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorLoggerSetMask", {"hiptensorLoggerSetMask", "", CONV_LIB_FUNC, API_TENSOR, 2}}, + {"cutensorLoggerForceDisable", {"hiptensorLoggerForceDisable", "", CONV_LIB_FUNC, API_TENSOR, 2}}, }; - const std::map CUDA_TENSOR_FUNCTION_VER_MAP { - {"cutensorCreate", {CUTENSOR_1700, CUDA_0, CUDA_0 }}, - {"cutensorDestroy", {CUTENSOR_1700, CUDA_0, CUDA_0 }}, - {"cutensorHandleResizePlanCache", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorHandleWritePlanCacheToFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorHandleReadPlanCacheFromFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorWriteKernelCacheToFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorReadKernelCacheFromFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreateTensorDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorInitTensorDescriptor", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000}}, - {"cutensorDestroyTensorDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreateElementwiseTrinary", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorElementwiseTrinaryExecute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreateElementwiseBinary", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorElementwiseBinaryExecute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreatePermutation", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorPermutation", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000}}, - {"cutensorPermute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreateContraction", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorContraction", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000}}, - {"cutensorDestroyOperationDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorOperationDescriptorSetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorOperationDescriptorGetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreatePlanPreference", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorDestroyPlanPreference", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorPlanPreferenceSetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorPlanGetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorEstimateWorkspaceSize", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreatePlan", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorDestroyPlan", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorContract", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorCreateReduction", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorReduction", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000}}, - {"cutensorReduce", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, - {"cutensorGetErrorString", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, - {"cutensorGetVersion", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, - {"cutensorGetCudartVersion", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, - {"cutensorLoggerSetCallback", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, - {"cutensorLoggerSetFile", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, - {"cutensorLoggerOpenFile", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, - {"cutensorLoggerSetLevel", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, - {"cutensorLoggerSetMask", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, - {"cutensorLoggerForceDisable", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, + {"cutensorCreate", {CUTENSOR_1700, CUDA_0, CUDA_0 }}, + {"cutensorDestroy", {CUTENSOR_1700, CUDA_0, CUDA_0 }}, + {"cutensorHandleResizePlanCache", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorHandleWritePlanCacheToFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorHandleReadPlanCacheFromFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorWriteKernelCacheToFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorReadKernelCacheFromFile", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreateTensorDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorInitTensorDescriptor", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"cutensorDestroyTensorDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreateElementwiseTrinary", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorElementwiseTrinaryExecute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreateElementwiseBinary", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorElementwiseBinaryExecute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreatePermutation", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorPermutation", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"cutensorPermute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreateContraction", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorContraction", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"cutensorDestroyOperationDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorOperationDescriptorSetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorOperationDescriptorGetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreatePlanPreference", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorDestroyPlanPreference", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorPlanPreferenceSetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorPlanGetAttribute", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorEstimateWorkspaceSize", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreatePlan", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorDestroyPlan", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorContract", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCreateReduction", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorReduction", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"cutensorReduce", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorGetErrorString", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorGetVersion", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorGetCudartVersion", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorLoggerSetCallback", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, + {"cutensorLoggerSetFile", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, + {"cutensorLoggerOpenFile", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, + {"cutensorLoggerSetLevel", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, + {"cutensorLoggerSetMask", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, + {"cutensorLoggerForceDisable", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, }; const std::map HIP_TENSOR_FUNCTION_VER_MAP { - {"hiptensorCreate", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorDestroy", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorInitTensorDescriptor", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorPermutation", {HIP_6010, HIP_0, HIP_0, }}, - {"hiptensorContraction", {HIP_6010, HIP_0, HIP_0, }}, - {"hiptensorReduction", {HIP_6030, HIP_0, HIP_0, }}, - {"hiptensorGetErrorString", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorGetHiprtVersion", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerSetCallback", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerSetFile", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerOpenFile", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerSetLevel", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerSetMask", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerForceDisable", {HIP_5070, HIP_0, HIP_0, }}, + {"hiptensorCreate", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorDestroy", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorInitTensorDescriptor", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorPermutation", {HIP_6010, HIP_0, HIP_0 }}, + {"hiptensorContraction", {HIP_6010, HIP_0, HIP_0 }}, + {"hiptensorReduction", {HIP_6030, HIP_0, HIP_0 }}, + {"hiptensorGetErrorString", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorGetHiprtVersion", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerSetCallback", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerSetFile", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerOpenFile", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerSetLevel", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerSetMask", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerForceDisable", {HIP_5070, HIP_0, HIP_0 }}, }; const std::map CUDA_TENSOR_API_SECTION_MAP { diff --git a/src/CUDA2HIP_TENSOR_API_types.cpp b/src/CUDA2HIP_TENSOR_API_types.cpp index de0f266f..59fb34d4 100644 --- a/src/CUDA2HIP_TENSOR_API_types.cpp +++ b/src/CUDA2HIP_TENSOR_API_types.cpp @@ -25,356 +25,355 @@ THE SOFTWARE. // Map of all functions const std::map CUDA_TENSOR_TYPE_NAME_MAP { // cuTENSOR enums - {"cutensorDataType_t", {"hiptensorComputeType_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"CUTENSOR_R_16F", {"HIPTENSOR_COMPUTE_16F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_16F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_16BF", {"HIPTENSOR_COMPUTE_16BF", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_16BF", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_32F", {"HIPTENSOR_COMPUTE_32F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_32F", {"HIPTENSOR_COMPUTE_C32F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_R_64F", {"HIPTENSOR_COMPUTE_64F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_64F", {"HIPTENSOR_COMPUTE_C64F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_R_4I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_4I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_4U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_4U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_8I", {"HIPTENSOR_COMPUTE_8I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_8I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_8U", {"HIPTENSOR_COMPUTE_8U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_8U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_16I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_16I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_16U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_16U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_32I", {"HIPTENSOR_COMPUTE_32I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_32I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_32U", {"HIPTENSOR_COMPUTE_32U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_C_32U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_64I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_64I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_64U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_64U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorDataType_t", {"hiptensorComputeType_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"CUTENSOR_R_16F", {"HIPTENSOR_COMPUTE_16F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_16F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_16BF", {"HIPTENSOR_COMPUTE_16BF", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_16BF", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_32F", {"HIPTENSOR_COMPUTE_32F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_32F", {"HIPTENSOR_COMPUTE_C32F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_R_64F", {"HIPTENSOR_COMPUTE_64F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_64F", {"HIPTENSOR_COMPUTE_C64F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_R_4I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_4I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_4U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_4U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_8I", {"HIPTENSOR_COMPUTE_8I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_8I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_8U", {"HIPTENSOR_COMPUTE_8U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_8U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_16I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_16I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_16U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_16U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_32I", {"HIPTENSOR_COMPUTE_32I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_32I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_32U", {"HIPTENSOR_COMPUTE_32U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_C_32U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_64I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_64I", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_64U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_64U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorComputeType_t", {"hiptensorComputeType_t", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_16F", {"HIPTENSOR_COMPUTE_16F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_16BF", {"HIPTENSOR_COMPUTE_16BF", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_TF32", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_COMPUTE_32F", {"HIPTENSOR_COMPUTE_32F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_64F", {"HIPTENSOR_COMPUTE_64F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_8U", {"HIPTENSOR_COMPUTE_8U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_8I", {"HIPTENSOR_COMPUTE_8I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_32U", {"HIPTENSOR_COMPUTE_32U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_COMPUTE_32I", {"HIPTENSOR_COMPUTE_32I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_R_MIN_16F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_MIN_16F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_MIN_32F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_MIN_32F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_MIN_64F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_MIN_64F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_MIN_8U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_MIN_32U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_MIN_16BF", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_R_MIN_TF32", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_C_MIN_TF32", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorComputeType_t", {"hiptensorComputeType_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_16F", {"HIPTENSOR_COMPUTE_16F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_16BF", {"HIPTENSOR_COMPUTE_16BF", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_TF32", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_COMPUTE_32F", {"HIPTENSOR_COMPUTE_32F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_64F", {"HIPTENSOR_COMPUTE_64F", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_8U", {"HIPTENSOR_COMPUTE_8U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_8I", {"HIPTENSOR_COMPUTE_8I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_32U", {"HIPTENSOR_COMPUTE_32U", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_COMPUTE_32I", {"HIPTENSOR_COMPUTE_32I", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_R_MIN_16F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_MIN_16F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_MIN_32F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_MIN_32F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_MIN_64F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_MIN_64F", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_MIN_8U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_MIN_32U", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_MIN_16BF", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_R_MIN_TF32", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_C_MIN_TF32", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorOperator_t", {"hiptensorOperator_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"CUTENSOR_OP_IDENTITY", {"HIPTENSOR_OP_IDENTITY", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_OP_SQRT", {"HIPTENSOR_OP_SQRT", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_OP_RELU", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_CONJ", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_RCP", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_SIGMOID", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_TANH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_EXP", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_LOG", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ABS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_NEG", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_SIN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_COS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_TAN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_SINH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_COSH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ASIN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ACOS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ATAN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ASINH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ACOSH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ATANH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_CEIL", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_FLOOR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_MISH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_SWISH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_SOFT_PLUS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_SOFT_SIGN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OP_ADD", {"HIPTENSOR_OP_ADD", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_OP_MUL", {"HIPTENSOR_OP_MUL", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_OP_MAX", {"HIPTENSOR_OP_MAX", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_OP_MIN", {"HIPTENSOR_OP_MIN", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_OP_UNKNOWN", {"HIPTENSOR_OP_UNKNOWN", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"cutensorOperator_t", {"hiptensorOperator_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"CUTENSOR_OP_IDENTITY", {"HIPTENSOR_OP_IDENTITY", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_OP_SQRT", {"HIPTENSOR_OP_SQRT", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_OP_RELU", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_CONJ", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_RCP", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_SIGMOID", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_TANH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_EXP", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_LOG", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ABS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_NEG", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_SIN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_COS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_TAN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_SINH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_COSH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ASIN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ACOS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ATAN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ASINH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ACOSH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ATANH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_CEIL", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_FLOOR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_MISH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_SWISH", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_SOFT_PLUS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_SOFT_SIGN", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OP_ADD", {"HIPTENSOR_OP_ADD", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_OP_MUL", {"HIPTENSOR_OP_MUL", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_OP_MAX", {"HIPTENSOR_OP_MAX", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_OP_MIN", {"HIPTENSOR_OP_MIN", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_OP_UNKNOWN", {"HIPTENSOR_OP_UNKNOWN", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"cutensorStatus_t", {"hiptensorStatus_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_SUCCESS", {"HIPTENSOR_STATUS_SUCCESS", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_NOT_INITIALIZED", {"HIPTENSOR_STATUS_NOT_INITIALIZED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_ALLOC_FAILED", {"HIPTENSOR_STATUS_ALLOC_FAILED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_INVALID_VALUE", {"HIPTENSOR_STATUS_INVALID_VALUE", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_ARCH_MISMATCH", {"HIPTENSOR_STATUS_ARCH_MISMATCH", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_MAPPING_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_STATUS_EXECUTION_FAILED", {"HIPTENSOR_STATUS_EXECUTION_FAILED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_INTERNAL_ERROR", {"HIPTENSOR_STATUS_INTERNAL_ERROR", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_NOT_SUPPORTED", {"HIPTENSOR_STATUS_NOT_SUPPORTED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_LICENSE_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_STATUS_CUBLAS_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_STATUS_CUDA_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_STATUS_INSUFFICIENT_WORKSPACE", {"HIPTENSOR_STATUS_INSUFFICIENT_WORKSPACE", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_INSUFFICIENT_DRIVER", {"HIPTENSOR_STATUS_INSUFFICIENT_DRIVER", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_STATUS_IO_ERROR", {"HIPTENSOR_STATUS_IO_ERROR", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"cutensorStatus_t", {"hiptensorStatus_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_SUCCESS", {"HIPTENSOR_STATUS_SUCCESS", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_NOT_INITIALIZED", {"HIPTENSOR_STATUS_NOT_INITIALIZED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_ALLOC_FAILED", {"HIPTENSOR_STATUS_ALLOC_FAILED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_INVALID_VALUE", {"HIPTENSOR_STATUS_INVALID_VALUE", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_ARCH_MISMATCH", {"HIPTENSOR_STATUS_ARCH_MISMATCH", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_MAPPING_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_STATUS_EXECUTION_FAILED", {"HIPTENSOR_STATUS_EXECUTION_FAILED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_INTERNAL_ERROR", {"HIPTENSOR_STATUS_INTERNAL_ERROR", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_NOT_SUPPORTED", {"HIPTENSOR_STATUS_NOT_SUPPORTED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_LICENSE_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_STATUS_CUBLAS_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_STATUS_CUDA_ERROR", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_STATUS_INSUFFICIENT_WORKSPACE", {"HIPTENSOR_STATUS_INSUFFICIENT_WORKSPACE", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_INSUFFICIENT_DRIVER", {"HIPTENSOR_STATUS_INSUFFICIENT_DRIVER", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_STATUS_IO_ERROR", {"HIPTENSOR_STATUS_IO_ERROR", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"cutensorAlgo_t", {"hiptensorAlgo_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"CUTENSOR_ALGO_DEFAULT_PATIENT", {"HIPTENSOR_ALGO_DEFAULT_PATIENT", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_ALGO_GETT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_ALGO_TGETT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_ALGO_TTGT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_ALGO_DEFAULT", {"HIPTENSOR_ALGO_DEFAULT", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"cutensorAlgo_t", {"hiptensorAlgo_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"CUTENSOR_ALGO_DEFAULT_PATIENT", {"HIPTENSOR_ALGO_DEFAULT_PATIENT", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_ALGO_GETT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_ALGO_TGETT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_ALGO_TTGT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_ALGO_DEFAULT", {"HIPTENSOR_ALGO_DEFAULT", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"cutensorWorksizePreference_t", {"hiptensorWorksizePreference_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"CUTENSOR_WORKSPACE_MIN", {"HIPTENSOR_WORKSPACE_MIN", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_WORKSPACE_DEFAULT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_WORKSPACE_RECOMMENDED", {"HIPTENSOR_WORKSPACE_RECOMMENDED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"CUTENSOR_WORKSPACE_MAX", {"HIPTENSOR_WORKSPACE_MAX", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"cutensorWorksizePreference_t", {"hiptensorWorksizePreference_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"CUTENSOR_WORKSPACE_MIN", {"HIPTENSOR_WORKSPACE_MIN", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_WORKSPACE_DEFAULT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_WORKSPACE_RECOMMENDED", {"HIPTENSOR_WORKSPACE_RECOMMENDED", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, + {"CUTENSOR_WORKSPACE_MAX", {"HIPTENSOR_WORKSPACE_MAX", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1}}, - {"cutensorOperationDescriptorAttribute_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_TAG", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_SCALAR_TYPE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_FLOPS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_MOVED_BYTES", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_LEFT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_RIGHT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_VALUE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorOperationDescriptorAttribute_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_TAG", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_SCALAR_TYPE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_FLOPS", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_MOVED_BYTES", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_LEFT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_RIGHT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_VALUE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorPlanPreferenceAttribute_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_PREFERENCE_AUTOTUNE_MODE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_PREFERENCE_CACHE_MODE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_PREFERENCE_INCREMENTAL_COUNT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_PREFERENCE_ALGO", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_PREFERENCE_KERNEL_RANK", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_PREFERENCE_JIT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorPlanPreferenceAttribute_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_PREFERENCE_AUTOTUNE_MODE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_PREFERENCE_CACHE_MODE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_PREFERENCE_INCREMENTAL_COUNT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_PREFERENCE_ALGO", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_PREFERENCE_KERNEL_RANK", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_PREFERENCE_JIT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorAutotuneMode_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_AUTOTUNE_MODE_NONE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_AUTOTUNE_MODE_INCREMENTAL", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorAutotuneMode_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_AUTOTUNE_MODE_NONE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_AUTOTUNE_MODE_INCREMENTAL", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorJitMode_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_JIT_MODE_NONE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_JIT_MODE_DEFAULT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorJitMode_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_JIT_MODE_NONE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_JIT_MODE_DEFAULT", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorCacheMode_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_CACHE_MODE_NONE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_CACHE_MODE_PEDANTIC", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorCacheMode_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_CACHE_MODE_NONE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_CACHE_MODE_PEDANTIC", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorPlanAttribute_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"CUTENSOR_PLAN_REQUIRED_WORKSPACE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, - - {"cutensorHandle_t", {"hiptensorHandle_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"cutensorHandle", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorTensorDescriptor_t", {"hiptensorTensorDescriptor_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"cutensorTensorDescriptor", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorContractionPlan_t", {"hiptensorContractionPlan_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"cutensorPlan_t", {"hiptensorContractionPlan_t", "", CONV_TYPE, API_TENSOR, 1}}, - {"cutensorPlan", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, - {"cutensorLoggerCallback_t", {"hiptensorLoggerCallback_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"cutensorPlanAttribute_t", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"CUTENSOR_PLAN_REQUIRED_WORKSPACE", {"", "", CONV_NUMERIC_LITERAL, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorHandle_t", {"hiptensorHandle_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"cutensorHandle", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorTensorDescriptor_t", {"hiptensorTensorDescriptor_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"cutensorTensorDescriptor", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorContractionPlan_t", {"hiptensorContractionPlan_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"cutensorPlan_t", {"hiptensorContractionPlan_t", "", CONV_TYPE, API_TENSOR, 1}}, + {"cutensorPlan", {"", "", CONV_TYPE, API_TENSOR, 1, UNSUPPORTED}}, + {"cutensorLoggerCallback_t", {"hiptensorLoggerCallback_t", "", CONV_TYPE, API_TENSOR, 1}}, }; const std::map CUDA_TENSOR_TYPE_NAME_VER_MAP { - {"cutensorDataType_t", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_16F", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_16F", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_16BF", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_16BF", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_32F", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_32F", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_64F", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_64F", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_4I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_4I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_4U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_4U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_8I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_8I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_8U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_8U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_16I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_16I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_16U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_16U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_32I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_32I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_32U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_32U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_64I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_64I", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_R_64U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_C_64U", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_COMPUTE_16F", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_16BF", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_TF32", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_32F", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_64F", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_8U", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_8I", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_32U", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_COMPUTE_32I", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_16F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_C_MIN_16F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_32F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_C_MIN_32F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_64F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_C_MIN_64F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_8U", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_32U", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_16BF", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_R_MIN_TF32", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"CUTENSOR_C_MIN_TF32", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000, }}, - {"cutensorOperator_t", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_IDENTITY", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SQRT", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_RELU", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_CONJ", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_RCP", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SIGMOID", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_TANH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_EXP", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_LOG", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ABS", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_NEG", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SIN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_COS", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_TAN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SINH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_COSH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ASIN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ACOS", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ATAN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ASINH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ACOSH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ATANH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_CEIL", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_FLOOR", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_MISH", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SWISH", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SOFT_PLUS", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_SOFT_SIGN", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_ADD", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_MUL", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_MAX", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_MIN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OP_UNKNOWN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"cutensorStatus_t", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_SUCCESS", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_NOT_INITIALIZED", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_ALLOC_FAILED", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_INVALID_VALUE", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_ARCH_MISMATCH", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_MAPPING_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_EXECUTION_FAILED", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_INTERNAL_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_NOT_SUPPORTED", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_LICENSE_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_CUBLAS_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_CUDA_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_INSUFFICIENT_WORKSPACE", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_INSUFFICIENT_DRIVER", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_STATUS_IO_ERROR", {CUTENSOR_1200, CUDA_0, CUDA_0, }}, - {"cutensorAlgo_t", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_ALGO_DEFAULT_PATIENT", {CUTENSOR_1400, CUDA_0, CUDA_0, }}, - {"CUTENSOR_ALGO_GETT", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_ALGO_TGETT", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_ALGO_TTGT", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_ALGO_DEFAULT", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"cutensorWorksizePreference_t", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_WORKSPACE_MIN", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"CUTENSOR_WORKSPACE_DEFAULT", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_WORKSPACE_RECOMMENDED", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_WORKSPACE_MAX", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"cutensorOperationDescriptorAttribute_t", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_TAG", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_SCALAR_TYPE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_FLOPS", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_MOVED_BYTES", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_LEFT", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_RIGHT", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_VALUE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorPlanPreferenceAttribute_t", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_PREFERENCE_AUTOTUNE_MODE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_PREFERENCE_CACHE_MODE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_PREFERENCE_INCREMENTAL_COUNT", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_PREFERENCE_ALGO", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_PREFERENCE_KERNEL_RANK", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_PREFERENCE_JIT", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorAutotuneMode_t", {CUTENSOR_1200, CUDA_0, CUDA_0, }}, - {"CUTENSOR_AUTOTUNE_MODE_NONE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_AUTOTUNE_MODE_INCREMENTAL", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_AUTOTUNE_NONE", {CUTENSOR_1200, CUDA_0, CUTENSOR_2000, }}, - {"CUTENSOR_AUTOTUNE_INCREMENTAL", {CUTENSOR_1200, CUDA_0, CUTENSOR_2000, }}, - {"cutensorJitMode_t", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_JIT_MODE_NONE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_JIT_MODE_DEFAULT", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorCacheMode_t", {CUTENSOR_1200, CUDA_0, CUDA_0, }}, - {"CUTENSOR_CACHE_MODE_NONE", {CUTENSOR_1200, CUDA_0, CUDA_0, }}, - {"CUTENSOR_CACHE_MODE_PEDANTIC", {CUTENSOR_1200, CUDA_0, CUDA_0, }}, - {"cutensorPlanAttribute_t", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"CUTENSOR_PLAN_REQUIRED_WORKSPACE", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorHandle_t", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"cutensorHandle", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorTensorDescriptor_t", {CUTENSOR_1010, CUDA_0, CUDA_0, }}, - {"cutensorTensorDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorContractionPlan_t", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000, }}, - {"cutensorPlan_t", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorPlan", {CUTENSOR_2000, CUDA_0, CUDA_0, }}, - {"cutensorLoggerCallback_t", {CUTENSOR_1320, CUDA_0, CUDA_0, }}, + {"cutensorDataType_t", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_16F", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_16F", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_16BF", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_16BF", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_32F", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_32F", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_64F", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_64F", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_4I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_4I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_4U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_4U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_8I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_8I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_8U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_8U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_16I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_16I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_16U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_16U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_32I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_32I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_32U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_32U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_64I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_64I", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_R_64U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_C_64U", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_COMPUTE_16F", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_16BF", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_TF32", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_32F", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_64F", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_8U", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_8I", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_32U", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_COMPUTE_32I", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_16F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_C_MIN_16F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_32F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_C_MIN_32F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_64F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_C_MIN_64F", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_8U", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_32U", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_16BF", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_R_MIN_TF32", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"CUTENSOR_C_MIN_TF32", {CUTENSOR_1010, CUTENSOR_1200, CUTENSOR_2000 }}, + {"cutensorOperator_t", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_IDENTITY", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SQRT", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_RELU", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_CONJ", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_RCP", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SIGMOID", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_TANH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_EXP", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_LOG", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ABS", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_NEG", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SIN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_COS", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_TAN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SINH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_COSH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ASIN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ACOS", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ATAN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ASINH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ACOSH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ATANH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_CEIL", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_FLOOR", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_MISH", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SWISH", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SOFT_PLUS", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_SOFT_SIGN", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_ADD", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_MUL", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_MAX", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_MIN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OP_UNKNOWN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorStatus_t", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_SUCCESS", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_NOT_INITIALIZED", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_ALLOC_FAILED", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_INVALID_VALUE", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_ARCH_MISMATCH", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_MAPPING_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_EXECUTION_FAILED", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_INTERNAL_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_NOT_SUPPORTED", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_LICENSE_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_CUBLAS_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_CUDA_ERROR", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_INSUFFICIENT_WORKSPACE", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_INSUFFICIENT_DRIVER", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_STATUS_IO_ERROR", {CUTENSOR_1200, CUDA_0, CUDA_0 }}, + {"cutensorAlgo_t", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_ALGO_DEFAULT_PATIENT", {CUTENSOR_1400, CUDA_0, CUDA_0 }}, + {"CUTENSOR_ALGO_GETT", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_ALGO_TGETT", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_ALGO_TTGT", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_ALGO_DEFAULT", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorWorksizePreference_t", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_WORKSPACE_MIN", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"CUTENSOR_WORKSPACE_DEFAULT", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_WORKSPACE_RECOMMENDED", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_WORKSPACE_MAX", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorOperationDescriptorAttribute_t", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_TAG", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_SCALAR_TYPE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_FLOPS", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_MOVED_BYTES", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_LEFT", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_RIGHT", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_OPERATION_DESCRIPTOR_PADDING_VALUE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorPlanPreferenceAttribute_t", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_PREFERENCE_AUTOTUNE_MODE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_PREFERENCE_CACHE_MODE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_PREFERENCE_INCREMENTAL_COUNT", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_PREFERENCE_ALGO", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_PREFERENCE_KERNEL_RANK", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_PREFERENCE_JIT", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorAutotuneMode_t", {CUTENSOR_1200, CUDA_0, CUDA_0 }}, + {"CUTENSOR_AUTOTUNE_MODE_NONE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_AUTOTUNE_MODE_INCREMENTAL", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_AUTOTUNE_NONE", {CUTENSOR_1200, CUDA_0, CUTENSOR_2000 }}, + {"CUTENSOR_AUTOTUNE_INCREMENTAL", {CUTENSOR_1200, CUDA_0, CUTENSOR_2000 }}, + {"cutensorJitMode_t", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_JIT_MODE_NONE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_JIT_MODE_DEFAULT", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorCacheMode_t", {CUTENSOR_1200, CUDA_0, CUDA_0 }}, + {"CUTENSOR_CACHE_MODE_NONE", {CUTENSOR_1200, CUDA_0, CUDA_0 }}, + {"CUTENSOR_CACHE_MODE_PEDANTIC", {CUTENSOR_1200, CUDA_0, CUDA_0 }}, + {"cutensorPlanAttribute_t", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"CUTENSOR_PLAN_REQUIRED_WORKSPACE", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorHandle_t", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorHandle", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorTensorDescriptor_t", {CUTENSOR_1010, CUDA_0, CUDA_0 }}, + {"cutensorTensorDescriptor", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorContractionPlan_t", {CUTENSOR_1010, CUDA_0, CUTENSOR_2000 }}, + {"cutensorPlan_t", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorPlan", {CUTENSOR_2000, CUDA_0, CUDA_0 }}, + {"cutensorLoggerCallback_t", {CUTENSOR_1320, CUDA_0, CUDA_0 }}, }; const std::map HIP_TENSOR_TYPE_NAME_VER_MAP { - {"hiptensorComputeType_t", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_16F", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_16BF", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_32F", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_C32F", {HIP_6010, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_64F", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_C64F", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_8I", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_8U", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_32I", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_COMPUTE_32U", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorOperator_t", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_IDENTITY", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_SQRT", {HIP_6020, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_ADD", {HIP_6030, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_MUL", {HIP_6030, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_MAX", {HIP_6030, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_MIN", {HIP_6030, HIP_0, HIP_0, }}, - {"HIPTENSOR_OP_UNKNOWN", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorStatus_t", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_SUCCESS", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_NOT_INITIALIZED", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_ALLOC_FAILED", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_INVALID_VALUE", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_ARCH_MISMATCH", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_EXECUTION_FAILED", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_INTERNAL_ERROR", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_NOT_SUPPORTED", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_INSUFFICIENT_WORKSPACE", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_INSUFFICIENT_DRIVER", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_STATUS_IO_ERROR", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorAlgo_t", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_ALGO_DEFAULT", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_ALGO_DEFAULT_PATIENT", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorWorksizePreference_t", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_WORKSPACE_MIN", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_WORKSPACE_RECOMMENDED", {HIP_5070, HIP_0, HIP_0, }}, - {"HIPTENSOR_WORKSPACE_MAX", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorHandle_t", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorTensorDescriptor_t", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorContractionPlan_t", {HIP_5070, HIP_0, HIP_0, }}, - {"hiptensorLoggerCallback_t", {HIP_5070, HIP_0, HIP_0, }}, + {"hiptensorComputeType_t", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_16F", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_16BF", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_32F", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_C32F", {HIP_6010, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_64F", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_C64F", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_8I", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_8U", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_32I", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_COMPUTE_32U", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorOperator_t", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_IDENTITY", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_SQRT", {HIP_6020, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_ADD", {HIP_6030, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_MUL", {HIP_6030, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_MAX", {HIP_6030, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_MIN", {HIP_6030, HIP_0, HIP_0 }}, + {"HIPTENSOR_OP_UNKNOWN", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorStatus_t", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_SUCCESS", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_NOT_INITIALIZED", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_ALLOC_FAILED", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_INVALID_VALUE", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_ARCH_MISMATCH", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_EXECUTION_FAILED", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_INTERNAL_ERROR", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_NOT_SUPPORTED", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_INSUFFICIENT_WORKSPACE", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_INSUFFICIENT_DRIVER", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_STATUS_IO_ERROR", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorAlgo_t", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_ALGO_DEFAULT", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_ALGO_DEFAULT_PATIENT", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorWorksizePreference_t", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_WORKSPACE_MIN", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_WORKSPACE_RECOMMENDED", {HIP_5070, HIP_0, HIP_0 }}, + {"HIPTENSOR_WORKSPACE_MAX", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorHandle_t", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorTensorDescriptor_t", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorContractionPlan_t", {HIP_5070, HIP_0, HIP_0 }}, + {"hiptensorLoggerCallback_t", {HIP_5070, HIP_0, HIP_0 }}, };