Merge branch 'master' into support-malloc
This commit is contained in:
Maneesh Gupta
GitHub
@@ -0,0 +1,516 @@
|
||||
# CUBLAS API supported by HIP
|
||||
|
||||
## **1. CUBLAS Data types**
|
||||
|
||||
| **type** | **CUDA** | **HIP** |**HIP value** (if differs) |
|
||||
|-------------:|---------------------------------------------------------------|------------------------------------------------------------|---------------------------|
|
||||
| enum |***`cublasStatus`*** |***`hipblasStatus_t`*** |
|
||||
| enum |***`cublasStatus_t`*** |***`hipblasStatus_t`*** |
|
||||
| 0 |*`CUBLAS_STATUS_SUCCESS`* |*`HIPBLAS_STATUS_SUCCESS`* |
|
||||
| 1 |*`CUBLAS_STATUS_NOT_INITIALIZED`* |*`HIPBLAS_STATUS_NOT_INITIALIZED`* |
|
||||
| 3 |*`CUBLAS_STATUS_ALLOC_FAILED`* |*`HIPBLAS_STATUS_ALLOC_FAILED`* | 2 |
|
||||
| 7 |*`CUBLAS_STATUS_INVALID_VALUE`* |*`HIPBLAS_STATUS_INVALID_VALUE`* | 3 |
|
||||
| 8 |*`CUBLAS_STATUS_ARCH_MISMATCH`* |*`HIPBLAS_STATUS_ARCH_MISMATCH`* | |
|
||||
| 11 |*`CUBLAS_STATUS_MAPPING_ERROR`* |*`HIPBLAS_STATUS_MAPPING_ERROR`* | 4 |
|
||||
| 13 |*`CUBLAS_STATUS_EXECUTION_FAILED`* |*`HIPBLAS_STATUS_EXECUTION_FAILED`* | 5 |
|
||||
| 14 |*`CUBLAS_STATUS_INTERNAL_ERROR`* |*`HIPBLAS_STATUS_INTERNAL_ERROR`* | 6 |
|
||||
| 15 |*`CUBLAS_STATUS_NOT_SUPPORTED`* |*`HIPBLAS_STATUS_NOT_SUPPORTED`* | 7 |
|
||||
| 16 |*`CUBLAS_STATUS_LICENSE_ERROR`* | |
|
||||
| enum |***`cublasOperation_t`*** |***`hipblasOperation_t`*** |
|
||||
| 0 |*`CUBLAS_OP_N`* |*`HIPBLAS_OP_N`* | 111 |
|
||||
| 1 |*`CUBLAS_OP_T`* |*`HIPBLAS_OP_T`* | 112 |
|
||||
| 2 |*`CUBLAS_OP_C`* |*`HIPBLAS_OP_C`* | 113 |
|
||||
| enum |***`cublasFillMode_t`*** |***`hipblasFillMode_t`*** |
|
||||
| 0 |*`CUBLAS_FILL_MODE_LOWER`* |*`HIPBLAS_FILL_MODE_LOWER`* | 121 |
|
||||
| 1 |*`CUBLAS_FILL_MODE_UPPER`* |*`HIPBLAS_FILL_MODE_UPPER`* | 122 |
|
||||
| enum |***`cublasDiagType_t`*** |***`hipblasDiagType_t`*** |
|
||||
| 0 |*`CUBLAS_DIAG_NON_UNIT`* |*`HIPBLAS_DIAG_NON_UNIT`* | 131 |
|
||||
| 1 |*`CUBLAS_DIAG_UNIT`* |*`HIPBLAS_DIAG_UNIT`* | 132 |
|
||||
| enum |***`cublasSideMode_t`*** |***`hipblasSideMode_t`*** |
|
||||
| 0 |*`CUBLAS_SIDE_LEFT`* |*`HIPBLAS_SIDE_LEFT`* | 141 |
|
||||
| 1 |*`CUBLAS_SIDE_RIGHT`* |*`HIPBLAS_SIDE_RIGHT`* | 142 |
|
||||
| enum |***`cublasPointerMode_t`*** |***`hipblasPointerMode_t`*** |
|
||||
| 0 |*`CUBLAS_POINTER_MODE_HOST`* |*`HIPBLAS_POINTER_MODE_HOST`* |
|
||||
| 1 |*`CUBLAS_POINTER_MODE_DEVICE`* |*`HIPBLAS_POINTER_MODE_DEVICE`* |
|
||||
| enum |***`cublasAtomicsMode_t`*** | |
|
||||
| 0 |*`CUBLAS_ATOMICS_NOT_ALLOWED`* | |
|
||||
| 1 |*`CUBLAS_ATOMICS_ALLOWED`* | |
|
||||
| enum |***`cublasAtomicsMode_t`*** | |
|
||||
| -1 |*`CUBLAS_GEMM_DFALT`* | |
|
||||
| -1 |*`CUBLAS_GEMM_DEFAULT`* | |
|
||||
| 0 |*`CUBLAS_GEMM_ALGO0`* | |
|
||||
| 1 |*`CUBLAS_GEMM_ALGO1`* | |
|
||||
| 2 |*`CUBLAS_GEMM_ALGO2`* | |
|
||||
| 3 |*`CUBLAS_GEMM_ALGO3`* | |
|
||||
| 4 |*`CUBLAS_GEMM_ALGO4`* | |
|
||||
| 5 |*`CUBLAS_GEMM_ALGO5`* | |
|
||||
| 6 |*`CUBLAS_GEMM_ALGO6`* | |
|
||||
| 7 |*`CUBLAS_GEMM_ALGO7`* | |
|
||||
| 8 |*`CUBLAS_GEMM_ALGO8`* | |
|
||||
| 9 |*`CUBLAS_GEMM_ALGO9`* | |
|
||||
| 10 |*`CUBLAS_GEMM_ALGO10`* | |
|
||||
| 11 |*`CUBLAS_GEMM_ALGO11`* | |
|
||||
| 12 |*`CUBLAS_GEMM_ALGO12`* | |
|
||||
| 13 |*`CUBLAS_GEMM_ALGO13`* | |
|
||||
| 14 |*`CUBLAS_GEMM_ALGO14`* | |
|
||||
| 15 |*`CUBLAS_GEMM_ALGO15`* | |
|
||||
| 16 |*`CUBLAS_GEMM_ALGO16`* | |
|
||||
| 17 |*`CUBLAS_GEMM_ALGO17`* | |
|
||||
| 18 |*`CUBLAS_GEMM_ALGO18`* | |
|
||||
| 19 |*`CUBLAS_GEMM_ALGO19`* | |
|
||||
| 20 |*`CUBLAS_GEMM_ALGO20`* | |
|
||||
| 21 |*`CUBLAS_GEMM_ALGO21`* | |
|
||||
| 22 |*`CUBLAS_GEMM_ALGO22`* | |
|
||||
| 23 |*`CUBLAS_GEMM_ALGO23`* | |
|
||||
| 99 |*`CUBLAS_GEMM_DEFAULT_TENSOR_OP`* | |
|
||||
| 99 |*`CUBLAS_GEMM_DFALT_TENSOR_OP`* | |
|
||||
| 100 |*`CUBLAS_GEMM_ALGO0_TENSOR_OP`* | |
|
||||
| 101 |*`CUBLAS_GEMM_ALGO1_TENSOR_OP`* | |
|
||||
| 102 |*`CUBLAS_GEMM_ALGO2_TENSOR_OP`* | |
|
||||
| 103 |*`CUBLAS_GEMM_ALGO3_TENSOR_OP`* | |
|
||||
| 104 |*`CUBLAS_GEMM_ALGO4_TENSOR_OP`* | |
|
||||
| 105 |*`CUBLAS_GEMM_ALGO5_TENSOR_OP`* | |
|
||||
| 106 |*`CUBLAS_GEMM_ALGO6_TENSOR_OP`* | |
|
||||
| 107 |*`CUBLAS_GEMM_ALGO7_TENSOR_OP`* | |
|
||||
| 108 |*`CUBLAS_GEMM_ALGO8_TENSOR_OP`* | |
|
||||
| 109 |*`CUBLAS_GEMM_ALGO9_TENSOR_OP`* | |
|
||||
| 110 |*`CUBLAS_GEMM_ALGO10_TENSOR_OP`* | |
|
||||
| 111 |*`CUBLAS_GEMM_ALGO11_TENSOR_OP`* | |
|
||||
| 112 |*`CUBLAS_GEMM_ALGO12_TENSOR_OP`* | |
|
||||
| 113 |*`CUBLAS_GEMM_ALGO13_TENSOR_OP`* | |
|
||||
| 114 |*`CUBLAS_GEMM_ALGO14_TENSOR_OP`* | |
|
||||
| 115 |*`CUBLAS_GEMM_ALGO15_TENSOR_OP`* | |
|
||||
| enum |***`cublasMath_t`*** | |
|
||||
| 0 |*`CUBLAS_DEFAULT_MATH`* | |
|
||||
| 1 |*`CUBLAS_TENSOR_OP_MATH`* | |
|
||||
| enum* |`cublasDataType_t` | |
|
||||
| struct |`cublasContext` | |
|
||||
| struct* |`cublasHandle_t` |`hipblasHandle_t` |
|
||||
|
||||
## **2. CUBLAS API functions**
|
||||
|
||||
| **CUDA** | **HIP** |
|
||||
|-----------------------------------------------------------|-------------------------------------------------|
|
||||
|`cublasCreate` |`hipblasCreate` |
|
||||
|`cublasCreate_v2` |`hipblasCreate` |
|
||||
|`cublasDestroy` |`hipblasDestroy` |
|
||||
|`cublasDestroy_v2` |`hipblasDestroy` |
|
||||
|`cublasGetVersion` | |
|
||||
|`cublasGetVersion_v2` | |
|
||||
|`cublasGetProperty` | |
|
||||
|`cublasGetStream` |`hipblasGetStream` |
|
||||
|`cublasGetStream_v2` |`hipblasGetStream_v2` |
|
||||
|`cublasSetStream` |`hipblasSetStream` |
|
||||
|`cublasSetStream_v2` |`hipblasSetStream_v2` |
|
||||
|`cublasGetPointerMode` |`hipblasGetPointerMode` |
|
||||
|`cublasGetPointerMode_v2` |`hipblasGetPointerMode_v2` |
|
||||
|`cublasSetPointerMode` |`hipblasSetPointerMode` |
|
||||
|`cublasSetPointerMode_v2` |`hipblasSetPointerMode_v2` |
|
||||
|`cublasGetAtomicsMode` | |
|
||||
|`cublasSetAtomicsMode` | |
|
||||
|`cublasGetMathMode` | |
|
||||
|`cublasSetMathMode` | |
|
||||
|`cublasLogCallback` | |
|
||||
|`cublasLoggerConfigure` | |
|
||||
|`cublasSetLoggerCallback` | |
|
||||
|`cublasGetLoggerCallback` | |
|
||||
|`cublasSetVector` |`hipblasSetVector` |
|
||||
|`cublasGetVector` |`hipblasGetVector` |
|
||||
|`cublasSetMatrix` |`hipblasSetMatrix` |
|
||||
|`cublasGetMatrix` |`hipblasGetMatrix` |
|
||||
|`cublasSetVectorAsync` | |
|
||||
|`cublasGetVectorAsync` | |
|
||||
|`cublasSetMatrixAsync` | |
|
||||
|`cublasGetMatrixAsync` | |
|
||||
|`cublasXerbla` | |
|
||||
|`cublasNrm2Ex` | |
|
||||
|`cublasSnrm2` |`hipblasSnrm2` |
|
||||
|`cublasSnrm2_v2` |`hipblasSnrm2` |
|
||||
|`cublasDnrm2` |`hipblasDnrm2` |
|
||||
|`cublasDnrm2_v2` |`hipblasDnrm2` |
|
||||
|`cublasScnrm2` | |
|
||||
|`cublasScnrm2_v2` | |
|
||||
|`cublasDznrm2` | |
|
||||
|`cublasDznrm2_v2` | |
|
||||
|`cublasDotEx` | |
|
||||
|`cublasDotcEx` | |
|
||||
|`cublasSdot` |`hipblasSdot` |
|
||||
|`cublasSdot_v2` |`hipblasSdot` |
|
||||
|`cublasDdot` |`hipblasDdot` |
|
||||
|`cublasDdot_v2` |`hipblasDdot` |
|
||||
|`cublasCdotu` | |
|
||||
|`cublasCdotu_v2` | |
|
||||
|`cublasCdotc` | |
|
||||
|`cublasCdotc_v2` | |
|
||||
|`cublasZdotu` | |
|
||||
|`cublasZdotu_v2` | |
|
||||
|`cublasZdotc` | |
|
||||
|`cublasZdotc_v2` | |
|
||||
|`cublasScalEx` | |
|
||||
|`cublasSscal` |`hipblasSscal` |
|
||||
|`cublasSscal_v2` |`hipblasSscal` |
|
||||
|`cublasDscal` |`hipblasDscal` |
|
||||
|`cublasDscal_v2` |`hipblasDscal` |
|
||||
|`cublasCscal` | |
|
||||
|`cublasCscal_v2` | |
|
||||
|`cublasCsscal` | |
|
||||
|`cublasCsscal_v2` | |
|
||||
|`cublasZscal` | |
|
||||
|`cublasZscal_v2` | |
|
||||
|`cublasZdscal` | |
|
||||
|`cublasZdscal_v2` | |
|
||||
|`cublasAxpyEx` | |
|
||||
|`cublasSaxpy` |`hipblasSaxpy` |
|
||||
|`cublasSaxpy_v2` |`hipblasSaxpy` |
|
||||
|`cublasDaxpy` |`hipblasDaxpy` |
|
||||
|`cublasDaxpy_v2` |`hipblasDaxpy` |
|
||||
|`cublasCaxpy` | |
|
||||
|`cublasCaxpy_v2` | |
|
||||
|`cublasZaxpy` | |
|
||||
|`cublasZaxpy_v2` | |
|
||||
|`cublasScopy` |`hipblasScopy` |
|
||||
|`cublasScopy_v2` |`hipblasScopy` |
|
||||
|`cublasDcopy` |`hipblasDcopy` |
|
||||
|`cublasDcopy_v2` |`hipblasDcopy` |
|
||||
|`cublasCcopy` | |
|
||||
|`cublasCcopy_v2` | |
|
||||
|`cublasZcopy` | |
|
||||
|`cublasZcopy_v2` | |
|
||||
|`cublasSswap` | |
|
||||
|`cublasSswap_v2` | |
|
||||
|`cublasDswap` | |
|
||||
|`cublasDswap_v2` | |
|
||||
|`cublasCswap` | |
|
||||
|`cublasCswap_v2` | |
|
||||
|`cublasZswap` | |
|
||||
|`cublasZswap_v2` | |
|
||||
|`cublasIsamax` |`hipblasIsamax` |
|
||||
|`cublasIsamax_v2` |`hipblasIsamax` |
|
||||
|`cublasIdamax` |`hipblasIdamax` |
|
||||
|`cublasIdamax_v2` |`hipblasIdamax` |
|
||||
|`cublasIcamax` | |
|
||||
|`cublasIcamax_v2` | |
|
||||
|`cublasIzamax` | |
|
||||
|`cublasIzamax_v2` | |
|
||||
|`cublasIsamin` | |
|
||||
|`cublasIsamin_v2` | |
|
||||
|`cublasIdamin` | |
|
||||
|`cublasIdamin_v2` | |
|
||||
|`cublasIcamin` | |
|
||||
|`cublasIcamin_v2` | |
|
||||
|`cublasIzamin` | |
|
||||
|`cublasIzamin_v2` | |
|
||||
|`cublasSasum` |`hipblasSasum` |
|
||||
|`cublasSasum_v2` |`hipblasSasum` |
|
||||
|`cublasDasum` |`hipblasDasum` |
|
||||
|`cublasDasum_v2` |`hipblasDasum` |
|
||||
|`cublasScasum` | |
|
||||
|`cublasScasum_v2` | |
|
||||
|`cublasDzasum` | |
|
||||
|`cublasDzasum_v2` | |
|
||||
|`cublasSrot` | |
|
||||
|`cublasSrot_v2` | |
|
||||
|`cublasDrot` | |
|
||||
|`cublasDrot_v2` | |
|
||||
|`cublasCrot` | |
|
||||
|`cublasCrot_v2` | |
|
||||
|`cublasZrot` | |
|
||||
|`cublasZrot_v2` | |
|
||||
|`cublasZdrot` | |
|
||||
|`cublasZdrot_v2` | |
|
||||
|`cublasSrotg` | |
|
||||
|`cublasSrotg_v2` | |
|
||||
|`cublasDrotg` | |
|
||||
|`cublasDrotg_v2` | |
|
||||
|`cublasCrotg` | |
|
||||
|`cublasCrotg_v2` | |
|
||||
|`cublasZrotg` | |
|
||||
|`cublasZrotg_v2` | |
|
||||
|`cublasSrotm` | |
|
||||
|`cublasSrotm_v2` | |
|
||||
|`cublasDrotm` | |
|
||||
|`cublasDrotm_v2` | |
|
||||
|`cublasSrotmg` | |
|
||||
|`cublasSrotmg_v2` | |
|
||||
|`cublasDrotmg` | |
|
||||
|`cublasDrotmg_v2` | |
|
||||
|`cublasSgemv` |`hipblasSgemv` |
|
||||
|`cublasSgemv_v2` |`hipblasSgemv` |
|
||||
|`cublasDgemv` |`hipblasDgemv` |
|
||||
|`cublasDgemv_v2` |`hipblasDgemv` |
|
||||
|`cublasCgemv` | |
|
||||
|`cublasCgemv_v2` | |
|
||||
|`cublasZgemv` | |
|
||||
|`cublasZgemv_v2` | |
|
||||
|`cublasSgbmv` | |
|
||||
|`cublasSgbmv_v2` | |
|
||||
|`cublasDgbmv` | |
|
||||
|`cublasDgbmv_v2` | |
|
||||
|`cublasCgbmv` | |
|
||||
|`cublasCgbmv_v2` | |
|
||||
|`cublasZgbmv` | |
|
||||
|`cublasZgbmv_v2` | |
|
||||
|`cublasStrmv` | |
|
||||
|`cublasStrmv_v2` | |
|
||||
|`cublasDtrmv` | |
|
||||
|`cublasDtrmv_v2` | |
|
||||
|`cublasCtrmv` | |
|
||||
|`cublasCtrmv_v2` | |
|
||||
|`cublasZtrmv` | |
|
||||
|`cublasZtrmv_v2` | |
|
||||
|`cublasStbmv` | |
|
||||
|`cublasStbmv_v2` | |
|
||||
|`cublasDtbmv` | |
|
||||
|`cublasDtbmv_v2` | |
|
||||
|`cublasCtbmv` | |
|
||||
|`cublasCtbmv_v2` | |
|
||||
|`cublasZtbmv` | |
|
||||
|`cublasZtbmv_v2` | |
|
||||
|`cublasStpmv` | |
|
||||
|`cublasStpmv_v2` | |
|
||||
|`cublasDtpmv` | |
|
||||
|`cublasDtpmv_v2` | |
|
||||
|`cublasCtpmv` | |
|
||||
|`cublasCtpmv_v2` | |
|
||||
|`cublasZtpmv` | |
|
||||
|`cublasZtpmv_v2` | |
|
||||
|`cublasStrsv` | |
|
||||
|`cublasStrsv_v2` | |
|
||||
|`cublasDtrsv` | |
|
||||
|`cublasDtrsv_v2` | |
|
||||
|`cublasCtrsv` | |
|
||||
|`cublasCtrsv_v2` | |
|
||||
|`cublasZtrsv` | |
|
||||
|`cublasZtrsv_v2` | |
|
||||
|`cublasStpsv` | |
|
||||
|`cublasStpsv_v2` | |
|
||||
|`cublasDtpsv` | |
|
||||
|`cublasDtpsv_v2` | |
|
||||
|`cublasCtpsv` | |
|
||||
|`cublasCtpsv_v2` | |
|
||||
|`cublasZtpsv` | |
|
||||
|`cublasZtpsv_v2` | |
|
||||
|`cublasStbsv` | |
|
||||
|`cublasStbsv_v2` | |
|
||||
|`cublasDtbsv` | |
|
||||
|`cublasDtbsv_v2` | |
|
||||
|`cublasCtbsv` | |
|
||||
|`cublasCtbsv_v2` | |
|
||||
|`cublasZtbsv` | |
|
||||
|`cublasZtbsv_v2` | |
|
||||
|`cublasSsymv` | |
|
||||
|`cublasSsymv_v2` | |
|
||||
|`cublasDsymv` | |
|
||||
|`cublasDsymv_v2` | |
|
||||
|`cublasCsymv` | |
|
||||
|`cublasCsymv_v2` | |
|
||||
|`cublasZsymv` | |
|
||||
|`cublasZsymv_v2` | |
|
||||
|`cublasChemv` | |
|
||||
|`cublasChemv_v2` | |
|
||||
|`cublasZhemv` | |
|
||||
|`cublasZhemv_v2` | |
|
||||
|`cublasSsbmv` | |
|
||||
|`cublasSsbmv_v2` | |
|
||||
|`cublasDsbmv` | |
|
||||
|`cublasDsbmv_v2` | |
|
||||
|`cublasChbmv` | |
|
||||
|`cublasChbmv_v2` | |
|
||||
|`cublasZhbmv` | |
|
||||
|`cublasZhbmv_v2` | |
|
||||
|`cublasSspmv` | |
|
||||
|`cublasSspmv_v2` | |
|
||||
|`cublasDspmv` | |
|
||||
|`cublasDspmv_v2` | |
|
||||
|`cublasChpmv` | |
|
||||
|`cublasChpmv_v2` | |
|
||||
|`cublasZhpmv` | |
|
||||
|`cublasZhpmv_v2` | |
|
||||
|`cublasSger` |`hipblasSger` |
|
||||
|`cublasSger_v2` |`hipblasSger` |
|
||||
|`cublasDger` |`hipblasDger` |
|
||||
|`cublasDger_v2` |`hipblasDger` |
|
||||
|`cublasCgeru` | |
|
||||
|`cublasCgeru_v2` | |
|
||||
|`cublasCgerc` | |
|
||||
|`cublasCgerc_v2` | |
|
||||
|`cublasZgeru` | |
|
||||
|`cublasZgeru_v2` | |
|
||||
|`cublasZgerc` | |
|
||||
|`cublasZgerc_v2` | |
|
||||
|`cublasSsyr` | |
|
||||
|`cublasSsyr_v2` | |
|
||||
|`cublasDsyr` | |
|
||||
|`cublasDsyr_v2` | |
|
||||
|`cublasCsyr` | |
|
||||
|`cublasCsyr_v2` | |
|
||||
|`cublasZsyr` | |
|
||||
|`cublasZsyr_v2` | |
|
||||
|`cublasCher` | |
|
||||
|`cublasCher_v2` | |
|
||||
|`cublasZher` | |
|
||||
|`cublasZher_v2` | |
|
||||
|`cublasSspr` | |
|
||||
|`cublasSspr_v2` | |
|
||||
|`cublasDspr` | |
|
||||
|`cublasDspr_v2` | |
|
||||
|`cublasChpr` | |
|
||||
|`cublasChpr_v2` | |
|
||||
|`cublasZhpr` | |
|
||||
|`cublasZhpr_v2` | |
|
||||
|`cublasSsyr2` | |
|
||||
|`cublasSsyr2_v2` | |
|
||||
|`cublasDsyr2` | |
|
||||
|`cublasDsyr2_v2` | |
|
||||
|`cublasCsyr2` | |
|
||||
|`cublasCsyr2_v2` | |
|
||||
|`cublasZsyr2` | |
|
||||
|`cublasZsyr2_v2` | |
|
||||
|`cublasCher2` | |
|
||||
|`cublasCher2_v2` | |
|
||||
|`cublasZher2` | |
|
||||
|`cublasZher2_v2` | |
|
||||
|`cublasSspr2` | |
|
||||
|`cublasSspr2_v2` | |
|
||||
|`cublasDspr2` | |
|
||||
|`cublasDspr2_v2` | |
|
||||
|`cublasChpr2` | |
|
||||
|`cublasChpr2_v2` | |
|
||||
|`cublasZhpr2` | |
|
||||
|`cublasZhpr2_v2` | |
|
||||
|`cublasSgemm` |`hipblasSgemm` |
|
||||
|`cublasSgemm_v2` |`hipblasSgemm` |
|
||||
|`cublasDgemm` |`hipblasDgemm` |
|
||||
|`cublasDgemm_v2` |`hipblasDgemm` |
|
||||
|`cublasCgemm` |`hipblasCgemm` |
|
||||
|`cublasCgemm_v2` |`hipblasCgemm` |
|
||||
|`cublasCgemm3m` | |
|
||||
|`cublasCgemm3mEx` | |
|
||||
|`cublasZgemm` | |
|
||||
|`cublasZgemm_v2` | |
|
||||
|`cublasZgemm3m` | |
|
||||
|`cublasHgemm` |`hipblasHgemm` |
|
||||
|`cublasSgemmEx` | |
|
||||
|`cublasGemmEx` | |
|
||||
|`cublasCgemmEx` | |
|
||||
|`cublasUint8gemmBias` | |
|
||||
|`cublasSsyrk` | |
|
||||
|`cublasSsyrk_v2` | |
|
||||
|`cublasDsyrk` | |
|
||||
|`cublasDsyrk_v2` | |
|
||||
|`cublasCsyrk` | |
|
||||
|`cublasCsyrk_v2` | |
|
||||
|`cublasZsyrk` | |
|
||||
|`cublasZsyrk_v2` | |
|
||||
|`cublasCsyrkEx` | |
|
||||
|`cublasCsyrk3mEx` | |
|
||||
|`cublasCherk` | |
|
||||
|`cublasCherk_v2` | |
|
||||
|`cublasZherk` | |
|
||||
|`cublasZherk_v2` | |
|
||||
|`cublasCherkEx` | |
|
||||
|`cublasCherk3mEx` | |
|
||||
|`cublasSsyr2k` | |
|
||||
|`cublasSsyr2k_v2` | |
|
||||
|`cublasDsyr2k` | |
|
||||
|`cublasDsyr2k_v2` | |
|
||||
|`cublasCsyr2k` | |
|
||||
|`cublasCsyr2k_v2` | |
|
||||
|`cublasZsyr2k` | |
|
||||
|`cublasZsyr2k_v2` | |
|
||||
|`cublasCher2k` | |
|
||||
|`cublasCher2k_v2` | |
|
||||
|`cublasZher2k` | |
|
||||
|`cublasZher2k_v2` | |
|
||||
|`cublasSsyrkx` | |
|
||||
|`cublasDsyrkx` | |
|
||||
|`cublasCsyrkx` | |
|
||||
|`cublasZsyrkx` | |
|
||||
|`cublasCherkx` | |
|
||||
|`cublasZherkx` | |
|
||||
|`cublasSsymm` | |
|
||||
|`cublasSsymm_v2` | |
|
||||
|`cublasDsymm` | |
|
||||
|`cublasDsymm_v2` | |
|
||||
|`cublasCsymm` | |
|
||||
|`cublasCsymm_v2` | |
|
||||
|`cublasZsymm` | |
|
||||
|`cublasZsymm_v2` | |
|
||||
|`cublasChemm` | |
|
||||
|`cublasChemm_v2` | |
|
||||
|`cublasZhemm` | |
|
||||
|`cublasZhemm_v2` | |
|
||||
|`cublasStrsm` |`hipblasStrsm` |
|
||||
|`cublasStrsm_v2` |`hipblasStrsm` |
|
||||
|`cublasDtrsm` |`hipblasDtrsm` |
|
||||
|`cublasDtrsm_v2` |`hipblasDtrsm` |
|
||||
|`cublasCtrsm` | |
|
||||
|`cublasCtrsm_v2` | |
|
||||
|`cublasZtrsm` | |
|
||||
|`cublasZtrsm_v2` | |
|
||||
|`cublasStrmm` | |
|
||||
|`cublasStrmm_v2` | |
|
||||
|`cublasDtrmm` | |
|
||||
|`cublasDtrmm_v2` | |
|
||||
|`cublasCtrmm` | |
|
||||
|`cublasCtrmm_v2` | |
|
||||
|`cublasZtrmm` | |
|
||||
|`cublasZtrmm_v2` | |
|
||||
|`cublasHgemmBatched` | |
|
||||
|`cublasSgemmBatched` |`hipblasSgemmBatched` |
|
||||
|`cublasDgemmBatched` |`hipblasDgemmBatched` |
|
||||
|`cublasCgemmBatched` | |
|
||||
|`cublasCgemm3mBatched` | |
|
||||
|`cublasZgemmBatched` | |
|
||||
|`cublasGemmBatchedEx` | |
|
||||
|`cublasGemmStridedBatchedEx` | |
|
||||
|`cublasSgemmStridedBatched` | |
|
||||
|`cublasDgemmStridedBatched` | |
|
||||
|`cublasCgemmStridedBatched` | |
|
||||
|`cublasCgemm3mStridedBatched` | |
|
||||
|`cublasZgemmStridedBatched` | |
|
||||
|`cublasHgemmStridedBatched` | |
|
||||
|`cublasSgeam` |`hipblasSgeam` |
|
||||
|`cublasDgeam` |`hipblasDgeam` |
|
||||
|`cublasCgeam` | |
|
||||
|`cublasZgeam` | |
|
||||
|`cublasSgetrfBatched` | |
|
||||
|`cublasDgetrfBatched` | |
|
||||
|`cublasCgetrfBatched` | |
|
||||
|`cublasZgetrfBatched` | |
|
||||
|`cublasSgetriBatched` | |
|
||||
|`cublasDgetriBatched` | |
|
||||
|`cublasCgetriBatched` | |
|
||||
|`cublasZgetriBatched` | |
|
||||
|`cublasSgetrsBatched` | |
|
||||
|`cublasDgetrsBatched` | |
|
||||
|`cublasCgetrsBatched` | |
|
||||
|`cublasZgetrsBatched` | |
|
||||
|`cublasStrsmBatched` | |
|
||||
|`cublasDtrsmBatched` | |
|
||||
|`cublasCtrsmBatched` | |
|
||||
|`cublasZtrsmBatched` | |
|
||||
|`cublasSmatinvBatched` | |
|
||||
|`cublasDmatinvBatched` | |
|
||||
|`cublasCmatinvBatched` | |
|
||||
|`cublasZmatinvBatched` | |
|
||||
|`cublasSgeqrfBatched` | |
|
||||
|`cublasDgeqrfBatched` | |
|
||||
|`cublasCgeqrfBatched` | |
|
||||
|`cublasZgeqrfBatched` | |
|
||||
|`cublasSgelsBatched` | |
|
||||
|`cublasDgelsBatched` | |
|
||||
|`cublasCgelsBatched` | |
|
||||
|`cublasZgelsBatched` | |
|
||||
|`cublasSdgmm` | |
|
||||
|`cublasDdgmm` | |
|
||||
|`cublasCdgmm` | |
|
||||
|`cublasZdgmm` | |
|
||||
|`cublasStpttr` | |
|
||||
|`cublasDtpttr` | |
|
||||
|`cublasCtpttr` | |
|
||||
|`cublasZtpttr` | |
|
||||
|`cublasStrttp` | |
|
||||
|`cublasDtrttp` | |
|
||||
|`cublasCtrttp` | |
|
||||
|`cublasZtrttp` | |
|
||||
@@ -0,0 +1,81 @@
|
||||
# CUFFT API supported by HIP
|
||||
|
||||
## **1. CUFFT Data types**
|
||||
|
||||
| **type** | **CUDA** | **HIP** |**HIP value** (if differs) |
|
||||
|-------------:|---------------------------------------------------------------|------------------------------------------------------------|---------------------------|
|
||||
| enum |***`cufftResult_t`*** |***`hipfftResult_t`*** |
|
||||
| enum |***`cufftResult`*** |***`hipfftResult`*** |
|
||||
| 0x0 |*`CUFFT_SUCCESS`* |*`HIPFFT_SUCCESS`* | 0 |
|
||||
| 0x1 |*`CUFFT_INVALID_PLAN`* |*`HIPFFT_INVALID_PLAN`* | 1 |
|
||||
| 0x2 |*`CUFFT_ALLOC_FAILED`* |*`HIPFFT_ALLOC_FAILED`* | 2 |
|
||||
| 0x3 |*`CUFFT_INVALID_TYPE`* |*`HIPFFT_INVALID_TYPE`* | 3 |
|
||||
| 0x4 |*`CUFFT_INVALID_VALUE`* |*`HIPFFT_INVALID_VALUE`* | 4 |
|
||||
| 0x5 |*`CUFFT_INTERNAL_ERROR`* |*`HIPFFT_INTERNAL_ERROR`* | 5 |
|
||||
| 0x6 |*`CUFFT_EXEC_FAILED`* |*`HIPFFT_EXEC_FAILED`* | 6 |
|
||||
| 0x7 |*`CUFFT_SETUP_FAILED`* |*`HIPFFT_SETUP_FAILED`* | 7 |
|
||||
| 0x8 |*`CUFFT_I;NVALID_SIZE`* |*`HIPFFT_INVALID_SIZE`* | 8 |
|
||||
| 0x9 |*`CUFFT_UNALIGNED_DATA`* |*`HIPFFT_UNALIGNED_DATA`* | 9 |
|
||||
| 0xA |*`CUFFT_INCOMPLETE_PARAMETER_LIST`* |*`HIPFFT_INCOMPLETE_PARAMETER_LIST`* | 10 |
|
||||
| 0xB |*`CUFFT_INVALID_DEVICE`* |*`HIPFFT_INVALID_DEVICE`* | 11 |
|
||||
| 0xC |*`CUFFT_PARSE_ERROR`* |*`HIPFFT_PARSE_ERROR`* | 12 |
|
||||
| 0xD |*`CUFFT_NO_WORKSPACE`* |*`HIPFFT_NO_WORKSPACE`* | 13 |
|
||||
| 0xE |*`CUFFT_NOT_IMPLEMENTED`* |*`HIPFFT_NOT_IMPLEMENTED`* | 14 |
|
||||
| 0xF |*`CUFFT_LICENSE_ERROR`* | |
|
||||
| 0x10 |*`CUFFT_NOT_SUPPORTED`* |*`HIPFFT_NOT_SUPPORTED`* | 16 |
|
||||
| float |***`cufftReal`*** |***`hipfftReal`*** |
|
||||
| double |***`cufftDoubleReal`*** |***`hipfftDoubleReal`*** |
|
||||
| float2 |***`cufftComplex`*** |***`hipfftComplex`*** |
|
||||
| double2 |***`cufftDoubleComplex`*** |***`hipfftDoubleComplex`*** |
|
||||
| define |`CUFFT_FORWARD` |`HIPFFT_FORWARD` |
|
||||
| define |`CUFFT_INVERSE` |`HIPFFT_BACKWARD` |
|
||||
| enum |***`cufftType_t`*** |***`hipfftType_t`*** |
|
||||
| enum |***`cufftType`*** |***`hipfftType`*** |
|
||||
| 0x2a |*`CUFFT_R2C`* |*`HIPFFT_R2C`* |
|
||||
| 0x2c |*`CUFFT_C2R`* |*`HIPFFT_C2R`* |
|
||||
| 0x29 |*`CUFFT_C2C`* |*`HIPFFT_C2C`* |
|
||||
| 0x6a |*`CUFFT_D2Z`* |*`HIPFFT_D2Z`* |
|
||||
| 0x6c |*`CUFFT_Z2D`* |*`HIPFFT_Z2D`* |
|
||||
| 0x69 |*`CUFFT_Z2Z`* |*`HIPFFT_Z2Z`* |
|
||||
| enum |***`cufftCompatibility_t`*** | |
|
||||
| enum |***`cufftCompatibility`*** | |
|
||||
| 0x01 |*`CUFFT_COMPATIBILITY_FFTW_PADDING`* | |
|
||||
| define |`CUFFT_COMPATIBILITY_DEFAULT` | |
|
||||
| int |***`cufftHandle`*** |***`hipfftHandle`*** |
|
||||
|
||||
## **2. CUFFT API functions**
|
||||
|
||||
| **CUDA** | **HIP** |
|
||||
|-----------------------------------------------------------|-------------------------------------------------|
|
||||
|`cufftPlan1d` |`hipfftPlan1d` |
|
||||
|`cufftPlan2d` |`hipfftPlan2d` |
|
||||
|`cufftPlan3d` |`hipfftPlan3d` |
|
||||
|`cufftPlanMany` |`hipfftPlanMany` |
|
||||
|`cufftMakePlan1d` |`hipfftMakePlan1d` |
|
||||
|`cufftMakePlan2d` |`hipfftMakePlan2d` |
|
||||
|`cufftMakePlan3d` |`hipfftMakePlan3d` |
|
||||
|`cufftMakePlanMany` |`hipfftMakePlanMany` |
|
||||
|`cufftMakePlanMany64` |`hipfftMakePlanMany64` |
|
||||
|`cufftGetSizeMany64` |`hipfftGetSizeMany64` |
|
||||
|`cufftEstimate1d` |`hipfftEstimate1d` |
|
||||
|`cufftEstimate2d` |`hipfftEstimate2d` |
|
||||
|`cufftEstimate3d` |`hipfftEstimate3d` |
|
||||
|`cufftEstimateMany` |`hipfftEstimateMany` |
|
||||
|`cufftCreate` |`hipfftCreate` |
|
||||
|`cufftGetSize1d` |`hipfftGetSize1d` |
|
||||
|`cufftGetSize2d` |`hipfftGetSize2d` |
|
||||
|`cufftGetSize3d` |`hipfftGetSize3d` |
|
||||
|`cufftGetSizeMany` |`hipfftGetSizeMany` |
|
||||
|`cufftGetSize` |`hipfftGetSize` |
|
||||
|`cufftSetWorkArea` |`hipfftSetWorkArea` |
|
||||
|`cufftSetAutoAllocation` |`hipfftSetAutoAllocation` |
|
||||
|`cufftExecC2C` |`hipfftExecC2C` |
|
||||
|`cufftExecR2C` |`hipfftExecR2C` |
|
||||
|`cufftExecC2R` |`hipfftExecC2R` |
|
||||
|`cufftExecZ2Z` |`hipfftExecZ2Z` |
|
||||
|`cufftExecD2Z` |`hipfftExecD2Z` |
|
||||
|`cufftExecZ2D` |`hipfftExecZ2D` |
|
||||
|`cufftSetStream` |`hipfftSetStream` |
|
||||
|`cufftDestroy` |`hipfftDestroy` |
|
||||
|`cufftGetVersion` |`hipfftGetVersion` |
|
||||
|`cufftGetProperty` | |
|
||||
@@ -0,0 +1,37 @@
|
||||
# cuComplex API supported by HIP
|
||||
|
||||
## **1. cuComplex Data types**
|
||||
|
||||
| **type** | **CUDA** | **HIP** |**HIP value** (if differs) |
|
||||
|-------------:|---------------------------------------------------------------|------------------------------------------------------------|---------------------------|
|
||||
| float2 |***`cuFloatComplex`*** |***`hipFloatComplex`*** | struct |
|
||||
| double2 |***`cuDoubleComplex`*** |***`hipDoubleComplex`*** | struct |
|
||||
| float2 |***`cuComplex`*** |***`hipComplex`*** | struct |
|
||||
|
||||
## **2. cuComplex API functions**
|
||||
|
||||
| **CUDA** | **HIP** |
|
||||
|-----------------------------------------------------------|-------------------------------------------------|
|
||||
|`cuCrealf` |`hipCrealf` |
|
||||
|`cuCimagf` |`hipCimagf` |
|
||||
|`make_cuFloatComplex` |`make_hipFloatComplex` |
|
||||
|`cuConjf` |`hipConjf` |
|
||||
|`cuCaddf` |`hipCaddf` |
|
||||
|`cuCsubf` |`hipCsubf` |
|
||||
|`cuCmulf` |`hipCmulf` |
|
||||
|`cuCdivf` |`hipCdivf` |
|
||||
|`cuCabsf` |`hipCabsf` |
|
||||
|`cuCreal` |`hipCreal` |
|
||||
|`cuCimag` |`hipCimag` |
|
||||
|`make_cuDoubleComplex` |`make_hipDoubleComplex` |
|
||||
|`cuConj` |`hipConj` |
|
||||
|`cuCadd` |`hipCadd` |
|
||||
|`cuCsub` |`hipCsub` |
|
||||
|`cuCmul` |`hipCmul` |
|
||||
|`cuCdiv` |`hipCdiv` |
|
||||
|`cuCabs` |`hipCabs` |
|
||||
|`make_cuComplex` |`make_hipComplex` |
|
||||
|`cuComplexFloatToDouble` |`hipComplexFloatToDouble` |
|
||||
|`cuComplexDoubleToFloat` |`hipComplexDoubleToFloat` |
|
||||
|`cuCfmaf` |`hipCfmaf` |
|
||||
|`cuCfma` |`hipCfma` |
|
||||
@@ -102,6 +102,15 @@ hip-clang links device code from different translation units together. For each
|
||||
hip-clang generates initializatiion and termination functions for each translation unit for host code compilation. The initialization functions call `__hipRegisterFatBinary` to register the fatbinary embeded in the ELF file. They also call `__hipRegisterFunction` and `__hipRegisterVar` to register kernel functions and device side global variables. The termination functions call `__hipUnregisterFatBinary`.
|
||||
hip-clang emits a global variable `__hip_gpubin_handle` of void** type with linkonce linkage and inital value 0 for each host translation unit. Each initialization function checks `__hip_gpubin_handle` and register the fatbinary only if `__hip_gpubin_handle` is 0 and saves the return value of `__hip_gpubin_handle` to `__hip_gpubin_handle`. This is to guarantee that the fatbinary is only registered once. Similar check is done in the termination functions.
|
||||
|
||||
#### Kernel Launching
|
||||
hip-clang supports kernel launching by CUDA `<<<>>>` syntax, hipLaunchKernel, and hipLaunchKernelGGL. The latter two are macros which expand to CUDA `<<<>>>` syntax.
|
||||
|
||||
In host code, hip-clang emits a stub function with the same name and arguments as the kernel. In the body of this function, hipSetupArgument is called for each kernel argument, then hipLaunchByPtr is called with a function pointer to the stub function.
|
||||
|
||||
When the executable or shared library is loaded by the dynamic linker, the initilization functions are called. In the initialization functions, when `__hipRegisterFatBinary` is called, the code objects containing all kernels are loaded; when `__hipRegisterFunction` is called, the stub functions are associated with the corresponding kernels in code objects.
|
||||
|
||||
In the host code, for the `<<<>>>` statement, hip-clang first emits call of hipConfigureCall to set up the threads and grids, then emits call of the stub function with the given arguments. In the stub function, when the runtime host API function hipLaunchByPtr is called, the real kernel associated with the stub function is launched.
|
||||
|
||||
### NVCC Implementation Notes
|
||||
|
||||
#### Interoperation between HIP and CUDA Driver
|
||||
|
||||
@@ -13,7 +13,7 @@ and provides practical suggestions on how to port CUDA code and work through com
|
||||
* [Converting a project "in-place"](#converting-a-project-in-place)
|
||||
- [Distinguishing Compiler Modes](#distinguishing-compiler-modes)
|
||||
* [Identifying HIP Target Platform](#identifying-hip-target-platform)
|
||||
* [Identifying the Compiler: hcc or nvcc](#identifying-the-compiler-hcc-or-nvcc)
|
||||
* [Identifying the Compiler: hcc, hip-clang, or nvcc](#identifying-the-compiler-hcc-hip-clang-or-nvcc)
|
||||
* [Identifying Current Compilation Pass: Host or Device](#identifying-current-compilation-pass-host-or-device)
|
||||
* [Compiler Defines: Summary](#compiler-defines-summary)
|
||||
- [Identifying Architecture Features](#identifying-architecture-features)
|
||||
@@ -137,13 +137,18 @@ All HIP projects target either the hcc or nvcc platform. The platform affects wh
|
||||
Many projects use a mixture of an accelerator compiler (hcc or nvcc) and a standard compiler (e.g., g++). These defines are set for both accelerator and standard compilers and thus are often the best option when writing code that uses conditional compilation.
|
||||
|
||||
|
||||
### Identifying the Compiler: hcc or nvcc
|
||||
Often, its useful to know whether the underlying compiler is hcc or nvcc. This knowledge can guard platform-specific code (features that only work on the nvcc or hcc path but not both) or aid in platform-specific performance tuning.
|
||||
### Identifying the Compiler: hcc, hip-clang or nvcc
|
||||
Often, its useful to know whether the underlying compiler is hcc, hip-clang or nvcc. This knowledge can guard platform-specific code (features that only work on the nvcc, hip-clang or hcc path but not all) or aid in platform-specific performance tuning.
|
||||
|
||||
```
|
||||
#ifdef __HCC__
|
||||
// Compiled with hcc
|
||||
|
||||
```
|
||||
```
|
||||
#ifdef __HIP__
|
||||
// Compiled with hip-clang
|
||||
|
||||
```
|
||||
|
||||
```
|
||||
@@ -159,9 +164,9 @@ Often, its useful to know whether the underlying compiler is hcc or nvcc. This
|
||||
// Compiled with nvcc (Cuda language extensions enabled)
|
||||
```
|
||||
|
||||
hcc directly generates the host code (using the Clang x86 target) and passes the code to another host compiler. Thus, it lacks the equivalent of the \__CUDA_ACC define.
|
||||
hcc and hip-clang directly generates the host code (using the Clang x86 target) and passes the code to another host compiler. Thus, it lacks the equivalent of the \__CUDA_ACC define.
|
||||
|
||||
The macro `__HIPCC__` is set if either `__HCC__` or `__CUDACC__` is defined. This configuration is useful in determining when code is being compiled using an accelerator-enabled compiler (hcc or nvcc) as opposed to a standard host compiler (GCC, ICC, Clang, etc.).
|
||||
The macro `__HIPCC__` is set if either `__HCC__`, `__HIP__` or `__CUDACC__` is defined. This configuration is useful in determining when code is being compiled using an accelerator-enabled compiler (hcc or nvcc) as opposed to a standard host compiler (GCC, ICC, Clang, etc.).
|
||||
|
||||
### Identifying Current Compilation Pass: Host or Device
|
||||
|
||||
@@ -176,22 +181,25 @@ Unlike `__CUDA_ARCH__`, the `__HIP_DEVICE_COMPILE__` value is 1 or undefined, an
|
||||
|
||||
|
||||
### Compiler Defines: Summary
|
||||
|Define | hcc | nvcc | Other (GCC, ICC, Clang, etc.)
|
||||
|--- | --- | --- |---|
|
||||
|Define | hcc | hip-clang | nvcc | Other (GCC, ICC, Clang, etc.)
|
||||
|--- | --- | --- | --- |---|
|
||||
|HIP-related defines:|
|
||||
|`__HIP_PLATFORM_HCC___`| Defined | Undefined | Defined if targeting hcc platform; undefined otherwise |
|
||||
|`__HIP_PLATFORM_NVCC___`| Undefined | Defined | Defined if targeting nvcc platform; undefined otherwise |
|
||||
|`__HIP_DEVICE_COMPILE__` | 1 if compiling for device; undefined if compiling for host |1 if compiling for device; undefined if compiling for host | Undefined
|
||||
|`__HIPCC__` | Defined | Defined | Undefined
|
||||
|`__HIP_ARCH_*` | 0 or 1 depending on feature support (see below) | 0 or 1 depending on feature support (see below) | 0
|
||||
|`__HIP_PLATFORM_HCC___`| Defined | Defined | Undefined | Defined if targeting hcc platform; undefined otherwise |
|
||||
|`__HIP_PLATFORM_NVCC___`| Undefined | Undefined | Defined | Defined if targeting nvcc platform; undefined otherwise |
|
||||
|`__HIP_DEVICE_COMPILE__` | 1 if compiling for device; undefined if compiling for host | 1 if compiling for device; undefined if compiling for host |1 if compiling for device; undefined if compiling for host | Undefined
|
||||
|`__HIPCC__` | Defined | Defined | Defined | Undefined
|
||||
|`__HIP_ARCH_*` | 0 or 1 depending on feature support (see below) |0 or 1 depending on feature support (see below) | 0 or 1 depending on feature support (see below) | 0
|
||||
|nvcc-related defines:|
|
||||
|`__CUDACC__` | Undefined | Defined if source code is compiled by nvcc; undefined otherwise | Undefined
|
||||
|`__NVCC__` | Undefined | Defined | Undefined
|
||||
|`__CUDA_ARCH__` | Undefined | Unsigned representing compute capability (e.g., "130") if in device code; 0 if in host code | Undefined
|
||||
|`__CUDACC__` | Undefined | Undefined | Defined if source code is compiled by nvcc; undefined otherwise | Undefined
|
||||
|`__NVCC__` | Undefined | Undefined | Defined | Undefined
|
||||
|`__CUDA_ARCH__` | Undefined | Undefined | Unsigned representing compute capability (e.g., "130") if in device code; 0 if in host code | Undefined
|
||||
|hcc-related defines:|
|
||||
|`__HCC__` | Defined | Undefined | Undefined
|
||||
|`__HCC_ACCELERATOR__` | Nonzero if in device code; otherwise undefined | Undefined | Undefined
|
||||
|`__clang__` | Defined | Undefined | Defined if using Clang; otherwise undefined
|
||||
|`__HCC__` | Defined | Undefined | Undefined | Undefined
|
||||
|`__HCC_ACCELERATOR__` | Nonzero if in device code; otherwise undefined | Undefined | Undefined | Undefined
|
||||
|hip-clang-related defines:|
|
||||
|`__HIP__` | Undefined | Defined | Undefined | Undefined
|
||||
|hcc/hip-clang common defines:|
|
||||
|`__clang__` | Defined | Defined | Undefined | Defined if using Clang; otherwise undefined
|
||||
|
||||
|
||||
## Identifying Architecture Features
|
||||
|
||||
مرجع در شماره جدید
Block a user