Merge pull request #958 from emankov/HIPIFY

[HIPIFY][rocBLAS][5.7.0] Sync with `rocBLAS` 5.7.0
ROCm · Jul 17, 2023 · 6eccfb7 · 6eccfb7
2 parents 4580f50 + 8ed6d8a
commit 6eccfb7
Show file tree

Hide file tree

Showing 7 changed files with 62 additions and 31 deletions.
diff --git a/bin/hipify-perl b/bin/hipify-perl
@@ -1173,6 +1173,7 @@ sub rocSubstitutions {
     subst("cublasGemmEx", "rocblas_gemm_ex", "library");
     subst("cublasGemmStridedBatchedEx", "rocblas_gemm_strided_batched_ex", "library");
     subst("cublasGetAtomicsMode", "rocblas_get_atomics_mode", "library");
+    subst("cublasGetMathMode", "rocblas_get_math_mode", "library");
     subst("cublasGetMatrix", "rocblas_get_matrix", "library");
     subst("cublasGetMatrixAsync", "rocblas_get_matrix_async", "library");
     subst("cublasGetPointerMode", "rocblas_get_pointer_mode", "library");
@@ -1219,6 +1220,7 @@ sub rocSubstitutions {
     subst("cublasSdot", "rocblas_sdot", "library");
     subst("cublasSdot_v2", "rocblas_sdot", "library");
     subst("cublasSetAtomicsMode", "rocblas_set_atomics_mode", "library");
+    subst("cublasSetMathMode", "rocblas_set_math_mode", "library");
     subst("cublasSetMatrix", "rocblas_set_matrix", "library");
     subst("cublasSetMatrixAsync", "rocblas_set_matrix_async", "library");
     subst("cublasSetPointerMode", "rocblas_set_pointer_mode", "library");
@@ -1563,12 +1565,14 @@ sub rocSubstitutions {
     subst("cuDoubleComplex", "rocblas_double_complex", "type");
     subst("cuFloatComplex", "rocblas_float_complex", "type");
     subst("cublasAtomicsMode_t", "rocblas_atomics_mode", "type");
+    subst("cublasComputeType_t", "rocblas_computetype", "type");
     subst("cublasContext", "_rocblas_handle", "type");
     subst("cublasDataType_t", "rocblas_datatype", "type");
     subst("cublasDiagType_t", "rocblas_diagonal", "type");
     subst("cublasFillMode_t", "rocblas_fill", "type");
     subst("cublasGemmAlgo_t", "rocblas_gemm_algo", "type");
     subst("cublasHandle_t", "rocblas_handle", "type");
+    subst("cublasMath_t", "rocblas_math_mode", "type");
     subst("cublasOperation_t", "rocblas_operation", "type");
     subst("cublasPointerMode_t", "rocblas_pointer_mode", "type");
     subst("cublasSideMode_t", "rocblas_side", "type");
@@ -1656,6 +1660,8 @@ sub rocSubstitutions {
     subst("pruneInfo_t", "rocsparse_mat_info", "type");
     subst("CUBLAS_ATOMICS_ALLOWED", "rocblas_atomics_allowed", "numeric_literal");
     subst("CUBLAS_ATOMICS_NOT_ALLOWED", "rocblas_atomics_not_allowed", "numeric_literal");
+    subst("CUBLAS_COMPUTE_32F", "rocblas_compute_type_f32", "numeric_literal");
+    subst("CUBLAS_DEFAULT_MATH", "rocblas_default_math", "numeric_literal");
     subst("CUBLAS_DIAG_NON_UNIT", "rocblas_diagonal_non_unit", "numeric_literal");
     subst("CUBLAS_DIAG_UNIT", "rocblas_diagonal_unit", "numeric_literal");
     subst("CUBLAS_FILL_MODE_FULL", "rocblas_fill_full", "numeric_literal");
@@ -1680,6 +1686,7 @@ sub rocSubstitutions {
     subst("CUBLAS_STATUS_NOT_INITIALIZED", "rocblas_status_invalid_handle", "numeric_literal");
     subst("CUBLAS_STATUS_NOT_SUPPORTED", "rocblas_status_perf_degraded", "numeric_literal");
     subst("CUBLAS_STATUS_SUCCESS", "rocblas_status_success", "numeric_literal");
+    subst("CUBLAS_TF32_TENSOR_OP_MATH", "rocblas_xf32_xdl_math_op", "numeric_literal");
     subst("CUDA_C_16BF", "rocblas_datatype_bf16_c", "numeric_literal");
     subst("CUDA_C_16F", "rocblas_datatype_f16_c", "numeric_literal");
     subst("CUDA_C_32F", "rocblas_datatype_f32_c", "numeric_literal");
@@ -9578,7 +9585,6 @@ sub warnRocOnlyUnsupportedFunctions {
         "cublasSetSmCountTarget",
         "cublasSetMatrix_64",
         "cublasSetMatrixAsync_64",
-        "cublasSetMathMode",
         "cublasSetLoggerCallback",
         "cublasSetKernelStream",
         "cublasSdot_v2_64",
@@ -9602,7 +9608,6 @@ sub warnRocOnlyUnsupportedFunctions {
         "cublasRotEx_64",
         "cublasNrm2Ex_64",
         "cublasMigrateComputeType",
-        "cublasMath_t",
         "cublasLoggerConfigure",
         "cublasLogCallback",
         "cublasIzamin_v2_64",
@@ -9645,7 +9650,6 @@ sub warnRocOnlyUnsupportedFunctions {
         "cublasGetProperty",
         "cublasGetMatrix_64",
         "cublasGetMatrixAsync_64",
-        "cublasGetMathMode",
         "cublasGetLoggerCallback",
         "cublasGetError",
         "cublasGetCudartVersion",
@@ -9872,7 +9876,6 @@ sub warnRocOnlyUnsupportedFunctions {
         "CUDA_C_4I",
         "CUDA_C_16U",
         "CUDA_C_16I",
-        "CUBLAS_TF32_TENSOR_OP_MATH",
         "CUBLAS_TENSOR_OP_MATH",
         "CUBLAS_STATUS_LICENSE_ERROR",
         "CUBLAS_PEDANTIC_MATH",
@@ -9920,7 +9923,6 @@ sub warnRocOnlyUnsupportedFunctions {
         "CUBLAS_GEMM_ALGO1",
         "CUBLAS_GEMM_ALGO0_TENSOR_OP",
         "CUBLAS_GEMM_ALGO0",
-        "CUBLAS_DEFAULT_MATH",
         "CUBLAS_COMPUTE_64F_PEDANTIC",
         "CUBLAS_COMPUTE_64F",
         "CUBLAS_COMPUTE_32I_PEDANTIC",
@@ -9929,7 +9931,6 @@ sub warnRocOnlyUnsupportedFunctions {
         "CUBLAS_COMPUTE_32F_FAST_TF32",
         "CUBLAS_COMPUTE_32F_FAST_16F",
         "CUBLAS_COMPUTE_32F_FAST_16BF",
-        "CUBLAS_COMPUTE_32F",
         "CUBLAS_COMPUTE_16F_PEDANTIC",
         "CUBLAS_COMPUTE_16F"
     )

diff --git a/docs/tables/CUBLAS_API_supported_by_HIP.md b/docs/tables/CUBLAS_API_supported_by_HIP.md
@@ -182,7 +182,7 @@
 |`cublasSetAtomicsMode`| | | |`hipblasSetAtomicsMode`|3.10.0| | | |
 |`cublasSetKernelStream`| | | | | | | | |
 |`cublasSetLoggerCallback`|9.2| | | | | | | |
-|`cublasSetMathMode`| | | | | | | | |
+|`cublasSetMathMode`|9.0| | | | | | | |
 |`cublasSetMatrix`| | | |`hipblasSetMatrix`|1.8.2| | | |
 |`cublasSetMatrixAsync`| | | |`hipblasSetMatrixAsync`|3.7.0| | | |
 |`cublasSetMatrixAsync_64`|12.0| | | | | | | |

diff --git a/docs/tables/CUBLAS_API_supported_by_HIP_and_ROC.md b/docs/tables/CUBLAS_API_supported_by_HIP_and_ROC.md
@@ -8,7 +8,7 @@
 |`CUBLAS_ATOMICS_NOT_ALLOWED`| | | |`HIPBLAS_ATOMICS_NOT_ALLOWED`|3.10.0| | | |`rocblas_atomics_not_allowed`|3.8.0| | | |
 |`CUBLAS_COMPUTE_16F`|11.0| | | | | | | | | | | | |
 |`CUBLAS_COMPUTE_16F_PEDANTIC`|11.0| | | | | | | | | | | | |
-|`CUBLAS_COMPUTE_32F`|11.0| | | | | | | | | | | | |
+|`CUBLAS_COMPUTE_32F`|11.0| | | | | | | |`rocblas_compute_type_f32`|5.7.0| | | |
 |`CUBLAS_COMPUTE_32F_FAST_16BF`|11.0| | | | | | | | | | | | |
 |`CUBLAS_COMPUTE_32F_FAST_16F`|11.0| | | | | | | | | | | | |
 |`CUBLAS_COMPUTE_32F_FAST_TF32`|11.0| | | | | | | | | | | | |
@@ -17,7 +17,7 @@
 |`CUBLAS_COMPUTE_32I_PEDANTIC`|11.0| | | | | | | | | | | | |
 |`CUBLAS_COMPUTE_64F`|11.0| | | | | | | | | | | | |
 |`CUBLAS_COMPUTE_64F_PEDANTIC`|11.0| | | | | | | | | | | | |
-|`CUBLAS_DEFAULT_MATH`|9.0| | | | | | | | | | | | |
+|`CUBLAS_DEFAULT_MATH`|9.0| | | | | | | |`rocblas_default_math`|5.7.0| | | |
 |`CUBLAS_DIAG_NON_UNIT`| | | |`HIPBLAS_DIAG_NON_UNIT`|1.8.2| | | |`rocblas_diagonal_non_unit`|1.5.0| | | |
 |`CUBLAS_DIAG_UNIT`| | | |`HIPBLAS_DIAG_UNIT`|1.8.2| | | |`rocblas_diagonal_unit`|1.5.0| | | |
 |`CUBLAS_FILL_MODE_FULL`|10.1| | |`HIPBLAS_FILL_MODE_FULL`|1.8.2| | | |`rocblas_fill_full`|1.5.0| | | |
@@ -89,16 +89,16 @@
 |`CUBLAS_STATUS_NOT_SUPPORTED`| | | |`HIPBLAS_STATUS_NOT_SUPPORTED`|1.8.2| | | |`rocblas_status_perf_degraded`|3.5.0| | | |
 |`CUBLAS_STATUS_SUCCESS`| | | |`HIPBLAS_STATUS_SUCCESS`|1.8.2| | | |`rocblas_status_success`|1.5.0| | | |
 |`CUBLAS_TENSOR_OP_MATH`|9.0|11.0| | | | | | | | | | | |
-|`CUBLAS_TF32_TENSOR_OP_MATH`|11.0| | | | | | | | | | | | |
+|`CUBLAS_TF32_TENSOR_OP_MATH`|11.0| | | | | | | |`rocblas_xf32_xdl_math_op`|5.7.0| | | |
 |`cublasAtomicsMode_t`| | | |`hipblasAtomicsMode_t`|3.10.0| | | |`rocblas_atomics_mode`|3.8.0| | | |
-|`cublasComputeType_t`|11.0| | |`hipblasDatatype_t`|1.8.2| | | | | | | | |
+|`cublasComputeType_t`|11.0| | |`hipblasDatatype_t`|1.8.2| | | |`rocblas_computetype`|5.7.0| | | |
 |`cublasContext`| | | | | | | | |`_rocblas_handle`|1.5.0| | | |
 |`cublasDataType_t`|7.5| | |`hipblasDatatype_t`|1.8.2| | | |`rocblas_datatype`|1.8.2| | | |
 |`cublasDiagType_t`| | | |`hipblasDiagType_t`|1.8.2| | | |`rocblas_diagonal`|1.5.0| | | |
 |`cublasFillMode_t`| | | |`hipblasFillMode_t`|1.8.2| | | |`rocblas_fill`|1.5.0| | | |
 |`cublasGemmAlgo_t`|8.0| | |`hipblasGemmAlgo_t`|1.8.2| | | |`rocblas_gemm_algo`|1.8.2| | | |
 |`cublasHandle_t`| | | |`hipblasHandle_t`|3.0.0| | | |`rocblas_handle`|1.5.0| | | |
-|`cublasMath_t`|9.0| | | | | | | | | | | | |
+|`cublasMath_t`|9.0| | | | | | | |`rocblas_math_mode`|5.7.0| | | |
 |`cublasOperation_t`| | | |`hipblasOperation_t`|1.8.2| | | |`rocblas_operation`|1.5.0| | | |
 |`cublasPointerMode_t`| | | |`hipblasPointerMode_t`|1.8.2| | | |`rocblas_pointer_mode`|1.6.0| | | |
 |`cublasSideMode_t`| | | |`hipblasSideMode_t`|1.8.2| | | |`rocblas_side`|1.5.0| | | |
@@ -156,7 +156,7 @@
 |`cublasGetCudartVersion`|10.1| | | | | | | | | | | | |
 |`cublasGetError`| | | | | | | | | | | | | |
 |`cublasGetLoggerCallback`|9.2| | | | | | | | | | | | |
-|`cublasGetMathMode`|9.0| | | | | | | | | | | | |
+|`cublasGetMathMode`|9.0| | | | | | | |`rocblas_get_math_mode`|5.7.0| | | |
 |`cublasGetMatrix`| | | |`hipblasGetMatrix`|1.8.2| | | |`rocblas_get_matrix`|1.6.0| | | |
 |`cublasGetMatrixAsync`| | | |`hipblasGetMatrixAsync`|3.7.0| | | |`rocblas_get_matrix_async`|3.5.0| | | |
 |`cublasGetMatrixAsync_64`|12.0| | | | | | | | | | | | |
@@ -182,7 +182,7 @@
 |`cublasSetAtomicsMode`| | | |`hipblasSetAtomicsMode`|3.10.0| | | |`rocblas_set_atomics_mode`|3.8.0| | | |
 |`cublasSetKernelStream`| | | | | | | | | | | | | |
 |`cublasSetLoggerCallback`|9.2| | | | | | | | | | | | |
-|`cublasSetMathMode`| | | | | | | | | | | | | |
+|`cublasSetMathMode`|9.0| | | | | | | |`rocblas_set_math_mode`|5.7.0| | | |
 |`cublasSetMatrix`| | | |`hipblasSetMatrix`|1.8.2| | | |`rocblas_set_matrix`|1.6.0| | | |
 |`cublasSetMatrixAsync`| | | |`hipblasSetMatrixAsync`|3.7.0| | | |`rocblas_set_matrix_async`|3.5.0| | | |
 |`cublasSetMatrixAsync_64`|12.0| | | | | | | | | | | | |

diff --git a/docs/tables/CUBLAS_API_supported_by_ROC.md b/docs/tables/CUBLAS_API_supported_by_ROC.md
@@ -8,7 +8,7 @@
 |`CUBLAS_ATOMICS_NOT_ALLOWED`| | | |`rocblas_atomics_not_allowed`|3.8.0| | | |
 |`CUBLAS_COMPUTE_16F`|11.0| | | | | | | |
 |`CUBLAS_COMPUTE_16F_PEDANTIC`|11.0| | | | | | | |
-|`CUBLAS_COMPUTE_32F`|11.0| | | | | | | |
+|`CUBLAS_COMPUTE_32F`|11.0| | |`rocblas_compute_type_f32`|5.7.0| | | |
 |`CUBLAS_COMPUTE_32F_FAST_16BF`|11.0| | | | | | | |
 |`CUBLAS_COMPUTE_32F_FAST_16F`|11.0| | | | | | | |
 |`CUBLAS_COMPUTE_32F_FAST_TF32`|11.0| | | | | | | |
@@ -17,7 +17,7 @@
 |`CUBLAS_COMPUTE_32I_PEDANTIC`|11.0| | | | | | | |
 |`CUBLAS_COMPUTE_64F`|11.0| | | | | | | |
 |`CUBLAS_COMPUTE_64F_PEDANTIC`|11.0| | | | | | | |
-|`CUBLAS_DEFAULT_MATH`|9.0| | | | | | | |
+|`CUBLAS_DEFAULT_MATH`|9.0| | |`rocblas_default_math`|5.7.0| | | |
 |`CUBLAS_DIAG_NON_UNIT`| | | |`rocblas_diagonal_non_unit`|1.5.0| | | |
 |`CUBLAS_DIAG_UNIT`| | | |`rocblas_diagonal_unit`|1.5.0| | | |
 |`CUBLAS_FILL_MODE_FULL`|10.1| | |`rocblas_fill_full`|1.5.0| | | |
@@ -89,16 +89,16 @@
 |`CUBLAS_STATUS_NOT_SUPPORTED`| | | |`rocblas_status_perf_degraded`|3.5.0| | | |
 |`CUBLAS_STATUS_SUCCESS`| | | |`rocblas_status_success`|1.5.0| | | |
 |`CUBLAS_TENSOR_OP_MATH`|9.0|11.0| | | | | | |
-|`CUBLAS_TF32_TENSOR_OP_MATH`|11.0| | | | | | | |
+|`CUBLAS_TF32_TENSOR_OP_MATH`|11.0| | |`rocblas_xf32_xdl_math_op`|5.7.0| | | |
 |`cublasAtomicsMode_t`| | | |`rocblas_atomics_mode`|3.8.0| | | |
-|`cublasComputeType_t`|11.0| | | | | | | |
+|`cublasComputeType_t`|11.0| | |`rocblas_computetype`|5.7.0| | | |
 |`cublasContext`| | | |`_rocblas_handle`|1.5.0| | | |
 |`cublasDataType_t`|7.5| | |`rocblas_datatype`|1.8.2| | | |
 |`cublasDiagType_t`| | | |`rocblas_diagonal`|1.5.0| | | |
 |`cublasFillMode_t`| | | |`rocblas_fill`|1.5.0| | | |
 |`cublasGemmAlgo_t`|8.0| | |`rocblas_gemm_algo`|1.8.2| | | |
 |`cublasHandle_t`| | | |`rocblas_handle`|1.5.0| | | |
-|`cublasMath_t`|9.0| | | | | | | |
+|`cublasMath_t`|9.0| | |`rocblas_math_mode`|5.7.0| | | |
 |`cublasOperation_t`| | | |`rocblas_operation`|1.5.0| | | |
 |`cublasPointerMode_t`| | | |`rocblas_pointer_mode`|1.6.0| | | |
 |`cublasSideMode_t`| | | |`rocblas_side`|1.5.0| | | |
@@ -156,7 +156,7 @@
 |`cublasGetCudartVersion`|10.1| | | | | | | |
 |`cublasGetError`| | | | | | | | |
 |`cublasGetLoggerCallback`|9.2| | | | | | | |
-|`cublasGetMathMode`|9.0| | | | | | | |
+|`cublasGetMathMode`|9.0| | |`rocblas_get_math_mode`|5.7.0| | | |
 |`cublasGetMatrix`| | | |`rocblas_get_matrix`|1.6.0| | | |
 |`cublasGetMatrixAsync`| | | |`rocblas_get_matrix_async`|3.5.0| | | |
 |`cublasGetMatrixAsync_64`|12.0| | | | | | | |
@@ -182,7 +182,7 @@
 |`cublasSetAtomicsMode`| | | |`rocblas_set_atomics_mode`|3.8.0| | | |
 |`cublasSetKernelStream`| | | | | | | | |
 |`cublasSetLoggerCallback`|9.2| | | | | | | |
-|`cublasSetMathMode`| | | | | | | | |
+|`cublasSetMathMode`|9.0| | |`rocblas_set_math_mode`|5.7.0| | | |
 |`cublasSetMatrix`| | | |`rocblas_set_matrix`|1.6.0| | | |
 |`cublasSetMatrixAsync`| | | |`rocblas_set_matrix_async`|3.5.0| | | |
 |`cublasSetMatrixAsync_64`|12.0| | | | | | | |

diff --git a/src/CUDA2HIP_BLAS_API_functions.cpp b/src/CUDA2HIP_BLAS_API_functions.cpp
@@ -35,8 +35,8 @@ const std::map<llvm::StringRef, hipCounter> CUDA_BLAS_FUNCTION_MAP {
   {"cublasSetKernelStream",          {"hipblasSetKernelStream",          "",                                         CONV_LIB_FUNC, API_BLAS, 4, UNSUPPORTED}},
   {"cublasGetAtomicsMode",           {"hipblasGetAtomicsMode",           "rocblas_get_atomics_mode",                 CONV_LIB_FUNC, API_BLAS, 4}},
   {"cublasSetAtomicsMode",           {"hipblasSetAtomicsMode",           "rocblas_set_atomics_mode",                 CONV_LIB_FUNC, API_BLAS, 4}},
-  {"cublasGetMathMode",              {"hipblasGetMathMode",              "",                                         CONV_LIB_FUNC, API_BLAS, 4, UNSUPPORTED}},
-  {"cublasSetMathMode",              {"hipblasSetMathMode",              "",                                         CONV_LIB_FUNC, API_BLAS, 4, UNSUPPORTED}},
+  {"cublasGetMathMode",              {"hipblasGetMathMode",              "rocblas_get_math_mode",                    CONV_LIB_FUNC, API_BLAS, 4, HIP_UNSUPPORTED}},
+  {"cublasSetMathMode",              {"hipblasSetMathMode",              "rocblas_set_math_mode",                    CONV_LIB_FUNC, API_BLAS, 4, HIP_UNSUPPORTED}},
   {"cublasMigrateComputeType",       {"hipblasMigrateComputeType",       "",                                         CONV_LIB_FUNC, API_BLAS, 4, UNSUPPORTED}},
   {"cublasGetSmCountTarget",         {"hipblasGetSmCountTarget",         "",                                         CONV_LIB_FUNC, API_BLAS, 4, UNSUPPORTED}},
   {"cublasSetSmCountTarget",         {"hipblasSetSmCountTarget",         "",                                         CONV_LIB_FUNC, API_BLAS, 4, UNSUPPORTED}},
@@ -1075,6 +1075,7 @@ const std::map<llvm::StringRef, hipCounter> CUDA_BLAS_FUNCTION_MAP {
 
 const std::map<llvm::StringRef, cudaAPIversions> CUDA_BLAS_FUNCTION_VER_MAP {
   {"cublasGetMathMode",                          {CUDA_90,  CUDA_0, CUDA_0}},
+  {"cublasSetMathMode",                          {CUDA_90,  CUDA_0, CUDA_0}},
   {"cublasMigrateComputeType",                   {CUDA_110, CUDA_0, CUDA_0}},
   {"cublasLogCallback",                          {CUDA_92,  CUDA_0, CUDA_0}},
   {"cublasLoggerConfigure",                      {CUDA_92,  CUDA_0, CUDA_0}},
@@ -1962,6 +1963,8 @@ const std::map<llvm::StringRef, hipAPIversions> HIP_BLAS_FUNCTION_VER_MAP {
   {"hipblasDgelsBatched",                        {HIP_5040, HIP_0,    HIP_0   }},
   {"hipblasCgelsBatched",                        {HIP_5040, HIP_0,    HIP_0   }},
   {"hipblasZgelsBatched",                        {HIP_5040, HIP_0,    HIP_0   }},
+  {"rocblas_get_math_mode",                      {HIP_5070, HIP_0,    HIP_0   }},
+  {"rocblas_set_math_mode",                      {HIP_5070, HIP_0,    HIP_0   }},
 };
 
 const std::map<unsigned int, llvm::StringRef> CUDA_BLAS_API_SECTION_MAP {