nvmath-python 0.9.0__cp314-cp314-win_amd64.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- nvmath/__init__.pxd +0 -0
- nvmath/__init__.py +45 -0
- nvmath/_internal/__init__.py +0 -0
- nvmath/_internal/attribute_ifc_factory.py +330 -0
- nvmath/_internal/layout.py +70 -0
- nvmath/_internal/templates.py +383 -0
- nvmath/_utils.py +150 -0
- nvmath/bindings/__init__.py +51 -0
- nvmath/bindings/_internal/__init__.pxd +0 -0
- nvmath/bindings/_internal/__init__.py +0 -0
- nvmath/bindings/_internal/cublas.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cublas.pxd +530 -0
- nvmath/bindings/_internal/cublasLt.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cublasLt.pxd +59 -0
- nvmath/bindings/_internal/cublasMp.pxd +52 -0
- nvmath/bindings/_internal/cudss.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cudss.pxd +47 -0
- nvmath/bindings/_internal/cufft.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cufft.pxd +70 -0
- nvmath/bindings/_internal/cufftMp.pxd +77 -0
- nvmath/bindings/_internal/curand.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/curand.pxd +42 -0
- nvmath/bindings/_internal/cusolver.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cusolver.pxd +15 -0
- nvmath/bindings/_internal/cusolverDn.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cusolverDn.pxd +406 -0
- nvmath/bindings/_internal/cusparse.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cusparse.pxd +277 -0
- nvmath/bindings/_internal/cusparseLt.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cusparseLt.pxd +48 -0
- nvmath/bindings/_internal/cutensor.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/cutensor.pxd +58 -0
- nvmath/bindings/_internal/mathdx.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/mathdx.pxd +116 -0
- nvmath/bindings/_internal/nvshmem.pxd +29 -0
- nvmath/bindings/_internal/utils.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/_internal/utils.pxd +179 -0
- nvmath/bindings/_internal/utils.pyi +10 -0
- nvmath/bindings/cublas.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cublas.pxd +558 -0
- nvmath/bindings/cublas.pyi +746 -0
- nvmath/bindings/cublasLt.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cublasLt.pxd +109 -0
- nvmath/bindings/cublasLt.pyi +1337 -0
- nvmath/bindings/cublasMp.pxd +85 -0
- nvmath/bindings/cublasMp.pyi +219 -0
- nvmath/bindings/cudss.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cudss.pxd +86 -0
- nvmath/bindings/cudss.pyi +282 -0
- nvmath/bindings/cufft.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cufft.pxd +118 -0
- nvmath/bindings/cufft.pyi +241 -0
- nvmath/bindings/cufftMp.pxd +124 -0
- nvmath/bindings/cufftMp.pyi +260 -0
- nvmath/bindings/curand.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/curand.pxd +71 -0
- nvmath/bindings/curand.pyi +159 -0
- nvmath/bindings/cusolver.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cusolver.pxd +62 -0
- nvmath/bindings/cusolver.pyi +242 -0
- nvmath/bindings/cusolverDn.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cusolverDn.pxd +430 -0
- nvmath/bindings/cusolverDn.pyi +416 -0
- nvmath/bindings/cusparse.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cusparse.pxd +338 -0
- nvmath/bindings/cusparse.pyi +659 -0
- nvmath/bindings/cusparseLt.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cusparseLt.pxd +99 -0
- nvmath/bindings/cusparseLt.pyi +198 -0
- nvmath/bindings/cutensor.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cutensor.pxd +98 -0
- nvmath/bindings/cutensor.pyi +294 -0
- nvmath/bindings/cycublas.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycublas.pxd +684 -0
- nvmath/bindings/cycublasLt.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycublasLt.pxd +1055 -0
- nvmath/bindings/cycublasMp.pxd +183 -0
- nvmath/bindings/cycudss.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycudss.pxd +224 -0
- nvmath/bindings/cycufft.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycufft.pxd +326 -0
- nvmath/bindings/cycufftMp.pxd +334 -0
- nvmath/bindings/cycurand.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycurand.pxd +146 -0
- nvmath/bindings/cycusolver.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycusolver.pxd +154 -0
- nvmath/bindings/cycusolverDn.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycusolverDn.pxd +446 -0
- nvmath/bindings/cycusparse.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycusparse.pxd +470 -0
- nvmath/bindings/cycusparseLt.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycusparseLt.pxd +150 -0
- nvmath/bindings/cycutensor.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cycutensor.pxd +192 -0
- nvmath/bindings/cymathdx.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/cymathdx.pxd +552 -0
- nvmath/bindings/cynvshmem.pxd +126 -0
- nvmath/bindings/mathdx.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/mathdx.pxd +182 -0
- nvmath/bindings/mathdx.pyi +1244 -0
- nvmath/bindings/nvpl/__init__.pxd +0 -0
- nvmath/bindings/nvpl/__init__.py +13 -0
- nvmath/bindings/nvpl/_internal/__init__.pxd +0 -0
- nvmath/bindings/nvpl/_internal/__init__.py +0 -0
- nvmath/bindings/nvpl/_internal/blas.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/nvpl/_internal/blas.pxd +125 -0
- nvmath/bindings/nvpl/_internal/fft.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/nvpl/_internal/fft.pxd +36 -0
- nvmath/bindings/nvpl/blas.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/nvpl/blas.pxd +85 -0
- nvmath/bindings/nvpl/blas.pyi +122 -0
- nvmath/bindings/nvpl/cyblas.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/nvpl/cyblas.pxd +166 -0
- nvmath/bindings/nvpl/cyfft.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/nvpl/cyfft.pxd +92 -0
- nvmath/bindings/nvpl/fft.cp314-win_amd64.pyd +0 -0
- nvmath/bindings/nvpl/fft.pxd +100 -0
- nvmath/bindings/nvpl/fft.pyi +100 -0
- nvmath/bindings/nvshmem.pxd +54 -0
- nvmath/bindings/nvshmem.pyi +179 -0
- nvmath/device/__init__.py +21 -0
- nvmath/device/_deprecated.py +33 -0
- nvmath/device/common.py +313 -0
- nvmath/device/common_backend.py +131 -0
- nvmath/device/common_cuda.py +201 -0
- nvmath/device/common_numba.py +310 -0
- nvmath/device/common_opaque_tensor.py +201 -0
- nvmath/device/cublasdx.py +1805 -0
- nvmath/device/cublasdx_backend.py +807 -0
- nvmath/device/cublasdx_numba.py +1612 -0
- nvmath/device/cufftdx.py +510 -0
- nvmath/device/cufftdx_backend.py +196 -0
- nvmath/device/cufftdx_numba.py +196 -0
- nvmath/device/curand_kernel.py +9147 -0
- nvmath/device/cusolverdx.py +2690 -0
- nvmath/device/cusolverdx_backend.py +440 -0
- nvmath/device/cusolverdx_numba.py +1624 -0
- nvmath/device/llvm_array.py +29 -0
- nvmath/device/random.py +445 -0
- nvmath/device/random_helpers.py +23 -0
- nvmath/device/random_states.py +187 -0
- nvmath/device/types.py +147 -0
- nvmath/device/vector_types_numba.py +203 -0
- nvmath/distributed/__init__.py +205 -0
- nvmath/distributed/_internal/__init__.py +0 -0
- nvmath/distributed/_internal/nvshmem.py +302 -0
- nvmath/distributed/_internal/tensor_ifc.py +67 -0
- nvmath/distributed/_internal/tensor_ifc_cupy.py +62 -0
- nvmath/distributed/_internal/tensor_ifc_host_device.py +165 -0
- nvmath/distributed/_internal/tensor_ifc_numpy.py +41 -0
- nvmath/distributed/_internal/tensor_ifc_torch.py +141 -0
- nvmath/distributed/_internal/tensor_wrapper.py +78 -0
- nvmath/distributed/_utils.py +167 -0
- nvmath/distributed/distribution.py +770 -0
- nvmath/distributed/fft/__init__.py +7 -0
- nvmath/distributed/fft/_configuration.py +79 -0
- nvmath/distributed/fft/fft.py +2801 -0
- nvmath/distributed/linalg/__init__.py +12 -0
- nvmath/distributed/linalg/_internal/__init__.py +3 -0
- nvmath/distributed/linalg/_internal/epilog_protocol.py +496 -0
- nvmath/distributed/linalg/_internal/matmul_desc_ifc.py +28 -0
- nvmath/distributed/linalg/advanced/__init__.py +8 -0
- nvmath/distributed/linalg/advanced/_configuration.py +166 -0
- nvmath/distributed/linalg/advanced/matmulmod.py +2908 -0
- nvmath/distributed/process_group.py +408 -0
- nvmath/distributed/reshape/__init__.py +6 -0
- nvmath/distributed/reshape/_configuration.py +39 -0
- nvmath/distributed/reshape/reshape.py +1256 -0
- nvmath/fft/__init__.py +7 -0
- nvmath/fft/_configuration.py +208 -0
- nvmath/fft/_exec_utils.py +109 -0
- nvmath/fft/_helpers.py +237 -0
- nvmath/fft/fft.py +2903 -0
- nvmath/internal/__init__.pxd +3 -0
- nvmath/internal/__init__.py +10 -0
- nvmath/internal/_bindings.cp314-win_amd64.pyd +0 -0
- nvmath/internal/_bindings.pxd +12 -0
- nvmath/internal/_device_utils.py +54 -0
- nvmath/internal/enum_utils.py +142 -0
- nvmath/internal/formatters.py +87 -0
- nvmath/internal/mem_limit.py +51 -0
- nvmath/internal/memory.cp314-win_amd64.pyd +0 -0
- nvmath/internal/memory.pxd +10 -0
- nvmath/internal/memory.pyi +46 -0
- nvmath/internal/ndbuffer/__init__.pxd +3 -0
- nvmath/internal/ndbuffer/__init__.py +7 -0
- nvmath/internal/ndbuffer/copy_kernel/args.h +34 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/array_view.h +52 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/elementwise.h +68 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/grid_indexer.h +69 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/transposed.h +240 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/type_utils.h +39 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/utils.h +132 -0
- nvmath/internal/ndbuffer/copy_kernel/copy_kernel_impl/vec.h +159 -0
- nvmath/internal/ndbuffer/copy_kernel/elementwise.h +53 -0
- nvmath/internal/ndbuffer/copy_kernel/transposed.h +58 -0
- nvmath/internal/ndbuffer/copy_kernel.cp314-win_amd64.pyd +0 -0
- nvmath/internal/ndbuffer/copy_kernel.pxd +9 -0
- nvmath/internal/ndbuffer/copy_kernel.pyi +9 -0
- nvmath/internal/ndbuffer/data_layout.cp314-win_amd64.pyd +0 -0
- nvmath/internal/ndbuffer/data_layout.pxd +72 -0
- nvmath/internal/ndbuffer/data_layout.pyi +23 -0
- nvmath/internal/ndbuffer/jit.cp314-win_amd64.pyd +0 -0
- nvmath/internal/ndbuffer/jit.pxd +7 -0
- nvmath/internal/ndbuffer/jit.pyi +6 -0
- nvmath/internal/ndbuffer/ndbuffer.cp314-win_amd64.pyd +0 -0
- nvmath/internal/ndbuffer/ndbuffer.pxd +36 -0
- nvmath/internal/ndbuffer/ndbuffer.pyi +41 -0
- nvmath/internal/ndbuffer/package_utils.cp314-win_amd64.pyd +0 -0
- nvmath/internal/ndbuffer/package_utils.pxd +11 -0
- nvmath/internal/ndbuffer/package_utils.pyi +12 -0
- nvmath/internal/package_ifc.py +147 -0
- nvmath/internal/package_ifc_cuda.py +49 -0
- nvmath/internal/package_ifc_cupy.py +49 -0
- nvmath/internal/package_ifc_torch.py +31 -0
- nvmath/internal/package_wrapper.py +14 -0
- nvmath/internal/tensor_ifc.py +194 -0
- nvmath/internal/tensor_ifc_cupy.py +241 -0
- nvmath/internal/tensor_ifc_ndbuffer.py +153 -0
- nvmath/internal/tensor_ifc_numpy.py +183 -0
- nvmath/internal/tensor_ifc_torch.py +176 -0
- nvmath/internal/tensor_wrapper.py +157 -0
- nvmath/internal/typemaps.py +162 -0
- nvmath/internal/utils.py +771 -0
- nvmath/linalg/__init__.py +46 -0
- nvmath/linalg/_internal/__init__.py +3 -0
- nvmath/linalg/_internal/algo_cap_ifc.py +82 -0
- nvmath/linalg/_internal/algo_config_ifc.py +43 -0
- nvmath/linalg/_internal/batch.py +217 -0
- nvmath/linalg/_internal/enum_to_tuples.py +64 -0
- nvmath/linalg/_internal/epilog_protocol.py +766 -0
- nvmath/linalg/_internal/layout.py +624 -0
- nvmath/linalg/_internal/matmul_desc_ifc.py +28 -0
- nvmath/linalg/_internal/matmul_pref_ifc.py +27 -0
- nvmath/linalg/_internal/matrix_layout_ifc.py +26 -0
- nvmath/linalg/_internal/typemaps.py +144 -0
- nvmath/linalg/_internal/utils.py +134 -0
- nvmath/linalg/advanced/__init__.py +8 -0
- nvmath/linalg/advanced/_algorithmmod.py +149 -0
- nvmath/linalg/advanced/_configuration.py +349 -0
- nvmath/linalg/advanced/helpers/__init__.py +5 -0
- nvmath/linalg/advanced/helpers/matmul.py +1348 -0
- nvmath/linalg/advanced/matmulmod.py +3937 -0
- nvmath/linalg/generic/__init__.py +43 -0
- nvmath/linalg/generic/_configuration/__init__.py +37 -0
- nvmath/linalg/generic/_configuration/layout.py +263 -0
- nvmath/linalg/generic/_configuration/match.py +609 -0
- nvmath/linalg/generic/_configuration/qualifiers.py +494 -0
- nvmath/linalg/generic/_configuration/wrap.py +217 -0
- nvmath/linalg/generic/_dtype.py +15 -0
- nvmath/linalg/generic/matmulmod.py +1060 -0
- nvmath/memory.py +282 -0
- nvmath/sparse/__init__.py +38 -0
- nvmath/sparse/_internal/__init__.py +21 -0
- nvmath/sparse/_internal/common_utils.py +144 -0
- nvmath/sparse/_internal/cudss_config_ifc.py +741 -0
- nvmath/sparse/_internal/cudss_data_ifc.py +399 -0
- nvmath/sparse/_internal/cudss_utils.py +445 -0
- nvmath/sparse/_internal/cusparse_utils.py +382 -0
- nvmath/sparse/_internal/sparse_bsc_ifc.py +300 -0
- nvmath/sparse/_internal/sparse_bsr_ifc.py +303 -0
- nvmath/sparse/_internal/sparse_coo_ifc.py +254 -0
- nvmath/sparse/_internal/sparse_csc_ifc.py +266 -0
- nvmath/sparse/_internal/sparse_csr_ifc.py +266 -0
- nvmath/sparse/_internal/sparse_dia_ifc.py +239 -0
- nvmath/sparse/_internal/sparse_format_helpers.py +601 -0
- nvmath/sparse/_internal/sparse_tensor_ifc.py +121 -0
- nvmath/sparse/_internal/sparse_ust_ifc.py +146 -0
- nvmath/sparse/_internal/utils.py +56 -0
- nvmath/sparse/advanced/__init__.py +7 -0
- nvmath/sparse/advanced/_configuration.py +230 -0
- nvmath/sparse/advanced/direct_solver.py +1884 -0
- nvmath/sparse/generic/__init__.py +7 -0
- nvmath/sparse/generic/_configuration.py +129 -0
- nvmath/sparse/generic/_helpers.py +137 -0
- nvmath/sparse/generic/_thunks.py +21 -0
- nvmath/sparse/generic/matmulmod.py +2426 -0
- nvmath/sparse/ust/__init__.py +7 -0
- nvmath/sparse/ust/_converters.py +378 -0
- nvmath/sparse/ust/_drawer.py +552 -0
- nvmath/sparse/ust/_emitter.py +553 -0
- nvmath/sparse/ust/_jit.py +212 -0
- nvmath/sparse/ust/_kernel.py +346 -0
- nvmath/sparse/ust/_semiring.py +71 -0
- nvmath/sparse/ust/_utils.py +102 -0
- nvmath/sparse/ust/interfaces/__init__.py +0 -0
- nvmath/sparse/ust/interfaces/torch_interface.py +476 -0
- nvmath/sparse/ust/semiring_ops.py +71 -0
- nvmath/sparse/ust/tensor.py +1027 -0
- nvmath/sparse/ust/tensor_drawer.py +552 -0
- nvmath/sparse/ust/tensor_emitter.py +558 -0
- nvmath/sparse/ust/tensor_format.py +935 -0
- nvmath/sparse/ust/tensor_jit.py +212 -0
- nvmath/sparse/ust/tensor_kernel.py +348 -0
- nvmath/sparse/ust/tensor_utils.py +380 -0
- nvmath/tensor/__init__.py +6 -0
- nvmath/tensor/_configuration.py +95 -0
- nvmath/tensor/_internal/__init__.py +3 -0
- nvmath/tensor/_internal/cutensor_config_ifc.py +247 -0
- nvmath/tensor/_internal/cutensor_utils.py +162 -0
- nvmath/tensor/_internal/data.py +43 -0
- nvmath/tensor/_internal/einsum_parser.py +444 -0
- nvmath/tensor/_internal/typemaps.py +96 -0
- nvmath/tensor/contract.py +1861 -0
- nvmath_python-0.9.0.dist-info/METADATA +127 -0
- nvmath_python-0.9.0.dist-info/RECORD +309 -0
- nvmath_python-0.9.0.dist-info/WHEEL +5 -0
- nvmath_python-0.9.0.dist-info/licenses/LICENSE +177 -0
- nvmath_python-0.9.0.dist-info/top_level.txt +2 -0
|
@@ -0,0 +1,166 @@
|
|
|
1
|
+
# Copyright (c) 2024-2026, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
|
|
2
|
+
#
|
|
3
|
+
# SPDX-License-Identifier: Apache-2.0
|
|
4
|
+
#
|
|
5
|
+
# This code was automatically generated with version 25.5, generator version 0.3.1.dev1303+g031f1197f. Do not modify it directly.
|
|
6
|
+
# This layer exposes the C header to Cython as-is.
|
|
7
|
+
|
|
8
|
+
from libc.stdint cimport int64_t, int32_t
|
|
9
|
+
|
|
10
|
+
###############################################################################
|
|
11
|
+
# Types (structs, enums, ...)
|
|
12
|
+
###############################################################################
|
|
13
|
+
|
|
14
|
+
# enums
|
|
15
|
+
ctypedef enum CBLAS_ORDER "CBLAS_ORDER":
|
|
16
|
+
CblasRowMajor "CblasRowMajor" = 101
|
|
17
|
+
CblasColMajor "CblasColMajor" = 102
|
|
18
|
+
|
|
19
|
+
ctypedef enum CBLAS_TRANSPOSE "CBLAS_TRANSPOSE":
|
|
20
|
+
CblasNoTrans "CblasNoTrans" = 111
|
|
21
|
+
CblasTrans "CblasTrans" = 112
|
|
22
|
+
CblasConjTrans "CblasConjTrans" = 113
|
|
23
|
+
|
|
24
|
+
ctypedef enum CBLAS_UPLO "CBLAS_UPLO":
|
|
25
|
+
CblasUpper "CblasUpper" = 121
|
|
26
|
+
CblasLower "CblasLower" = 122
|
|
27
|
+
|
|
28
|
+
ctypedef enum CBLAS_DIAG "CBLAS_DIAG":
|
|
29
|
+
CblasNonUnit "CblasNonUnit" = 131
|
|
30
|
+
CblasUnit "CblasUnit" = 132
|
|
31
|
+
|
|
32
|
+
ctypedef enum CBLAS_SIDE "CBLAS_SIDE":
|
|
33
|
+
CblasLeft "CblasLeft" = 141
|
|
34
|
+
CblasRight "CblasRight" = 142
|
|
35
|
+
|
|
36
|
+
|
|
37
|
+
# types
|
|
38
|
+
ctypedef int64_t nvpl_int64_t 'nvpl_int64_t'
|
|
39
|
+
ctypedef int32_t nvpl_int32_t 'nvpl_int32_t'
|
|
40
|
+
ctypedef struct nvpl_scomplex_t 'nvpl_scomplex_t':
|
|
41
|
+
float real
|
|
42
|
+
float imag
|
|
43
|
+
|
|
44
|
+
ctypedef struct nvpl_dcomplex_t 'nvpl_dcomplex_t':
|
|
45
|
+
double real
|
|
46
|
+
double imag
|
|
47
|
+
|
|
48
|
+
ctypedef nvpl_int32_t nvpl_int_t 'nvpl_int_t'
|
|
49
|
+
|
|
50
|
+
|
|
51
|
+
###############################################################################
|
|
52
|
+
# Functions
|
|
53
|
+
###############################################################################
|
|
54
|
+
|
|
55
|
+
cdef int MKL_mkl_set_num_threads_local(int nth) except?-42 nogil
|
|
56
|
+
cdef void MKL_mkl_set_num_threads(int nth) except* nogil
|
|
57
|
+
cdef void openblas_openblas_set_num_threads(int num_threads) except* nogil
|
|
58
|
+
cdef int openblas_openblas_set_num_threads_local(int num_threads) except?-42 nogil
|
|
59
|
+
cdef int nvpl_blas_get_version() except?-42 nogil
|
|
60
|
+
cdef int nvpl_blas_get_max_threads() except?-42 nogil
|
|
61
|
+
cdef void nvpl_blas_set_num_threads(int nthr) except* nogil
|
|
62
|
+
cdef int nvpl_blas_set_num_threads_local(int nthr_local) except?-42 nogil
|
|
63
|
+
cdef void cblas_sgemv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const float alpha, const float* A, const nvpl_int_t lda, const float* X, const nvpl_int_t incX, const float beta, float* Y, const nvpl_int_t incY) except* nogil
|
|
64
|
+
cdef void cblas_sgbmv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t KL, const nvpl_int_t KU, const float alpha, const float* A, const nvpl_int_t lda, const float* X, const nvpl_int_t incX, const float beta, float* Y, const nvpl_int_t incY) except* nogil
|
|
65
|
+
cdef void cblas_strmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const float* A, const nvpl_int_t lda, float* X, const nvpl_int_t incX) except* nogil
|
|
66
|
+
cdef void cblas_stbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const float* A, const nvpl_int_t lda, float* X, const nvpl_int_t incX) except* nogil
|
|
67
|
+
cdef void cblas_stpmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const float* Ap, float* X, const nvpl_int_t incX) except* nogil
|
|
68
|
+
cdef void cblas_strsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const float* A, const nvpl_int_t lda, float* X, const nvpl_int_t incX) except* nogil
|
|
69
|
+
cdef void cblas_stbsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const float* A, const nvpl_int_t lda, float* X, const nvpl_int_t incX) except* nogil
|
|
70
|
+
cdef void cblas_stpsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const float* Ap, float* X, const nvpl_int_t incX) except* nogil
|
|
71
|
+
cdef void cblas_dgemv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const double alpha, const double* A, const nvpl_int_t lda, const double* X, const nvpl_int_t incX, const double beta, double* Y, const nvpl_int_t incY) except* nogil
|
|
72
|
+
cdef void cblas_dgbmv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t KL, const nvpl_int_t KU, const double alpha, const double* A, const nvpl_int_t lda, const double* X, const nvpl_int_t incX, const double beta, double* Y, const nvpl_int_t incY) except* nogil
|
|
73
|
+
cdef void cblas_dtrmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const double* A, const nvpl_int_t lda, double* X, const nvpl_int_t incX) except* nogil
|
|
74
|
+
cdef void cblas_dtbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const double* A, const nvpl_int_t lda, double* X, const nvpl_int_t incX) except* nogil
|
|
75
|
+
cdef void cblas_dtpmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const double* Ap, double* X, const nvpl_int_t incX) except* nogil
|
|
76
|
+
cdef void cblas_dtrsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const double* A, const nvpl_int_t lda, double* X, const nvpl_int_t incX) except* nogil
|
|
77
|
+
cdef void cblas_dtbsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const double* A, const nvpl_int_t lda, double* X, const nvpl_int_t incX) except* nogil
|
|
78
|
+
cdef void cblas_dtpsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const double* Ap, double* X, const nvpl_int_t incX) except* nogil
|
|
79
|
+
cdef void cblas_cgemv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
80
|
+
cdef void cblas_cgbmv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t KL, const nvpl_int_t KU, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
81
|
+
cdef void cblas_ctrmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
82
|
+
cdef void cblas_ctbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
83
|
+
cdef void cblas_ctpmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* Ap, void* X, const nvpl_int_t incX) except* nogil
|
|
84
|
+
cdef void cblas_ctrsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
85
|
+
cdef void cblas_ctbsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
86
|
+
cdef void cblas_ctpsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* Ap, void* X, const nvpl_int_t incX) except* nogil
|
|
87
|
+
cdef void cblas_zgemv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
88
|
+
cdef void cblas_zgbmv(const CBLAS_ORDER order, const CBLAS_TRANSPOSE TransA, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t KL, const nvpl_int_t KU, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
89
|
+
cdef void cblas_ztrmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
90
|
+
cdef void cblas_ztbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
91
|
+
cdef void cblas_ztpmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* Ap, void* X, const nvpl_int_t incX) except* nogil
|
|
92
|
+
cdef void cblas_ztrsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
93
|
+
cdef void cblas_ztbsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const nvpl_int_t K, const void* A, const nvpl_int_t lda, void* X, const nvpl_int_t incX) except* nogil
|
|
94
|
+
cdef void cblas_ztpsv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t N, const void* Ap, void* X, const nvpl_int_t incX) except* nogil
|
|
95
|
+
cdef void cblas_ssymv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const float* A, const nvpl_int_t lda, const float* X, const nvpl_int_t incX, const float beta, float* Y, const nvpl_int_t incY) except* nogil
|
|
96
|
+
cdef void cblas_ssbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const nvpl_int_t K, const float alpha, const float* A, const nvpl_int_t lda, const float* X, const nvpl_int_t incX, const float beta, float* Y, const nvpl_int_t incY) except* nogil
|
|
97
|
+
cdef void cblas_sspmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const float* Ap, const float* X, const nvpl_int_t incX, const float beta, float* Y, const nvpl_int_t incY) except* nogil
|
|
98
|
+
cdef void cblas_sger(const CBLAS_ORDER order, const nvpl_int_t M, const nvpl_int_t N, const float alpha, const float* X, const nvpl_int_t incX, const float* Y, const nvpl_int_t incY, float* A, const nvpl_int_t lda) except* nogil
|
|
99
|
+
cdef void cblas_ssyr(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const float* X, const nvpl_int_t incX, float* A, const nvpl_int_t lda) except* nogil
|
|
100
|
+
cdef void cblas_sspr(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const float* X, const nvpl_int_t incX, float* Ap) except* nogil
|
|
101
|
+
cdef void cblas_ssyr2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const float* X, const nvpl_int_t incX, const float* Y, const nvpl_int_t incY, float* A, const nvpl_int_t lda) except* nogil
|
|
102
|
+
cdef void cblas_sspr2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const float* X, const nvpl_int_t incX, const float* Y, const nvpl_int_t incY, float* A) except* nogil
|
|
103
|
+
cdef void cblas_dsymv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const double* A, const nvpl_int_t lda, const double* X, const nvpl_int_t incX, const double beta, double* Y, const nvpl_int_t incY) except* nogil
|
|
104
|
+
cdef void cblas_dsbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const nvpl_int_t K, const double alpha, const double* A, const nvpl_int_t lda, const double* X, const nvpl_int_t incX, const double beta, double* Y, const nvpl_int_t incY) except* nogil
|
|
105
|
+
cdef void cblas_dspmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const double* Ap, const double* X, const nvpl_int_t incX, const double beta, double* Y, const nvpl_int_t incY) except* nogil
|
|
106
|
+
cdef void cblas_dger(const CBLAS_ORDER order, const nvpl_int_t M, const nvpl_int_t N, const double alpha, const double* X, const nvpl_int_t incX, const double* Y, const nvpl_int_t incY, double* A, const nvpl_int_t lda) except* nogil
|
|
107
|
+
cdef void cblas_dsyr(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const double* X, const nvpl_int_t incX, double* A, const nvpl_int_t lda) except* nogil
|
|
108
|
+
cdef void cblas_dspr(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const double* X, const nvpl_int_t incX, double* Ap) except* nogil
|
|
109
|
+
cdef void cblas_dsyr2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const double* X, const nvpl_int_t incX, const double* Y, const nvpl_int_t incY, double* A, const nvpl_int_t lda) except* nogil
|
|
110
|
+
cdef void cblas_dspr2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const double* X, const nvpl_int_t incX, const double* Y, const nvpl_int_t incY, double* A) except* nogil
|
|
111
|
+
cdef void cblas_chemv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
112
|
+
cdef void cblas_chbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
113
|
+
cdef void cblas_chpmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* Ap, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
114
|
+
cdef void cblas_cgeru(const CBLAS_ORDER order, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* A, const nvpl_int_t lda) except* nogil
|
|
115
|
+
cdef void cblas_cgerc(const CBLAS_ORDER order, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* A, const nvpl_int_t lda) except* nogil
|
|
116
|
+
cdef void cblas_cher(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const void* X, const nvpl_int_t incX, void* A, const nvpl_int_t lda) except* nogil
|
|
117
|
+
cdef void cblas_chpr(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const float alpha, const void* X, const nvpl_int_t incX, void* A) except* nogil
|
|
118
|
+
cdef void cblas_cher2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* A, const nvpl_int_t lda) except* nogil
|
|
119
|
+
cdef void cblas_chpr2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* Ap) except* nogil
|
|
120
|
+
cdef void cblas_zhemv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
121
|
+
cdef void cblas_zhbmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
122
|
+
cdef void cblas_zhpmv(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* Ap, const void* X, const nvpl_int_t incX, const void* beta, void* Y, const nvpl_int_t incY) except* nogil
|
|
123
|
+
cdef void cblas_zgeru(const CBLAS_ORDER order, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* A, const nvpl_int_t lda) except* nogil
|
|
124
|
+
cdef void cblas_zgerc(const CBLAS_ORDER order, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* A, const nvpl_int_t lda) except* nogil
|
|
125
|
+
cdef void cblas_zher(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const void* X, const nvpl_int_t incX, void* A, const nvpl_int_t lda) except* nogil
|
|
126
|
+
cdef void cblas_zhpr(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const double alpha, const void* X, const nvpl_int_t incX, void* A) except* nogil
|
|
127
|
+
cdef void cblas_zher2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* A, const nvpl_int_t lda) except* nogil
|
|
128
|
+
cdef void cblas_zhpr2(const CBLAS_ORDER order, const CBLAS_UPLO Uplo, const nvpl_int_t N, const void* alpha, const void* X, const nvpl_int_t incX, const void* Y, const nvpl_int_t incY, void* Ap) except* nogil
|
|
129
|
+
cdef void cblas_sgemm(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const float alpha, const float* A, const nvpl_int_t lda, const float* B, const nvpl_int_t ldb, const float beta, float* C, const nvpl_int_t ldc) except* nogil
|
|
130
|
+
cdef void cblas_ssymm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const nvpl_int_t M, const nvpl_int_t N, const float alpha, const float* A, const nvpl_int_t lda, const float* B, const nvpl_int_t ldb, const float beta, float* C, const nvpl_int_t ldc) except* nogil
|
|
131
|
+
cdef void cblas_ssyrk(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const float alpha, const float* A, const nvpl_int_t lda, const float beta, float* C, const nvpl_int_t ldc) except* nogil
|
|
132
|
+
cdef void cblas_ssyr2k(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const float alpha, const float* A, const nvpl_int_t lda, const float* B, const nvpl_int_t ldb, const float beta, float* C, const nvpl_int_t ldc) except* nogil
|
|
133
|
+
cdef void cblas_strmm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const float alpha, const float* A, const nvpl_int_t lda, float* B, const nvpl_int_t ldb) except* nogil
|
|
134
|
+
cdef void cblas_strsm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const float alpha, const float* A, const nvpl_int_t lda, float* B, const nvpl_int_t ldb) except* nogil
|
|
135
|
+
cdef void cblas_dgemm(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const double alpha, const double* A, const nvpl_int_t lda, const double* B, const nvpl_int_t ldb, const double beta, double* C, const nvpl_int_t ldc) except* nogil
|
|
136
|
+
cdef void cblas_dsymm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const nvpl_int_t M, const nvpl_int_t N, const double alpha, const double* A, const nvpl_int_t lda, const double* B, const nvpl_int_t ldb, const double beta, double* C, const nvpl_int_t ldc) except* nogil
|
|
137
|
+
cdef void cblas_dsyrk(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const double alpha, const double* A, const nvpl_int_t lda, const double beta, double* C, const nvpl_int_t ldc) except* nogil
|
|
138
|
+
cdef void cblas_dsyr2k(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const double alpha, const double* A, const nvpl_int_t lda, const double* B, const nvpl_int_t ldb, const double beta, double* C, const nvpl_int_t ldc) except* nogil
|
|
139
|
+
cdef void cblas_dtrmm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const double alpha, const double* A, const nvpl_int_t lda, double* B, const nvpl_int_t ldb) except* nogil
|
|
140
|
+
cdef void cblas_dtrsm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const double alpha, const double* A, const nvpl_int_t lda, double* B, const nvpl_int_t ldb) except* nogil
|
|
141
|
+
cdef void cblas_cgemm(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
142
|
+
cdef void cblas_csymm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
143
|
+
cdef void cblas_csyrk(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
144
|
+
cdef void cblas_csyr2k(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
145
|
+
cdef void cblas_ctrmm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, void* B, const nvpl_int_t ldb) except* nogil
|
|
146
|
+
cdef void cblas_ctrsm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, void* B, const nvpl_int_t ldb) except* nogil
|
|
147
|
+
cdef void cblas_zgemm(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
148
|
+
cdef void cblas_zsymm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
149
|
+
cdef void cblas_zsyrk(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
150
|
+
cdef void cblas_zsyr2k(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
151
|
+
cdef void cblas_ztrmm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, void* B, const nvpl_int_t ldb) except* nogil
|
|
152
|
+
cdef void cblas_ztrsm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE TransA, const CBLAS_DIAG Diag, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, void* B, const nvpl_int_t ldb) except* nogil
|
|
153
|
+
cdef void cblas_chemm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
154
|
+
cdef void cblas_cherk(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const float alpha, const void* A, const nvpl_int_t lda, const float beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
155
|
+
cdef void cblas_cher2k(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const float beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
156
|
+
cdef void cblas_zhemm(const CBLAS_ORDER Order, const CBLAS_SIDE Side, const CBLAS_UPLO Uplo, const nvpl_int_t M, const nvpl_int_t N, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const void* beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
157
|
+
cdef void cblas_zherk(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const double alpha, const void* A, const nvpl_int_t lda, const double beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
158
|
+
cdef void cblas_zher2k(const CBLAS_ORDER Order, const CBLAS_UPLO Uplo, const CBLAS_TRANSPOSE Trans, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const void* B, const nvpl_int_t ldb, const double beta, void* C, const nvpl_int_t ldc) except* nogil
|
|
159
|
+
cdef void cblas_sgemm_batch(CBLAS_ORDER Order, CBLAS_TRANSPOSE* TransA_array, CBLAS_TRANSPOSE* TransB_array, nvpl_int_t* M_array, nvpl_int_t* N_array, nvpl_int_t* K_array, const float* alpha_array, const float** A_array, nvpl_int_t* lda_array, const float** B_array, nvpl_int_t* ldb_array, const float* beta_array, float** C_array, nvpl_int_t* ldc_array, nvpl_int_t group_count, nvpl_int_t* group_size) except* nogil
|
|
160
|
+
cdef void cblas_dgemm_batch(CBLAS_ORDER Order, CBLAS_TRANSPOSE* TransA_array, CBLAS_TRANSPOSE* TransB_array, nvpl_int_t* M_array, nvpl_int_t* N_array, nvpl_int_t* K_array, const double* alpha_array, const double** A_array, nvpl_int_t* lda_array, const double** B_array, nvpl_int_t* ldb_array, const double* beta_array, double** C_array, nvpl_int_t* ldc_array, nvpl_int_t group_count, nvpl_int_t* group_size) except* nogil
|
|
161
|
+
cdef void cblas_cgemm_batch(CBLAS_ORDER Order, CBLAS_TRANSPOSE* TransA_array, CBLAS_TRANSPOSE* TransB_array, nvpl_int_t* M_array, nvpl_int_t* N_array, nvpl_int_t* K_array, const void* alpha_array, const void** A_array, nvpl_int_t* lda_array, const void** B_array, nvpl_int_t* ldb_array, const void* beta_array, void** C_array, nvpl_int_t* ldc_array, nvpl_int_t group_count, nvpl_int_t* group_size) except* nogil
|
|
162
|
+
cdef void cblas_zgemm_batch(CBLAS_ORDER Order, CBLAS_TRANSPOSE* TransA_array, CBLAS_TRANSPOSE* TransB_array, nvpl_int_t* M_array, nvpl_int_t* N_array, nvpl_int_t* K_array, const void* alpha_array, const void** A_array, nvpl_int_t* lda_array, const void** B_array, nvpl_int_t* ldb_array, const void* beta_array, void** C_array, nvpl_int_t* ldc_array, nvpl_int_t group_count, nvpl_int_t* group_size) except* nogil
|
|
163
|
+
cdef void cblas_sgemm_batch_strided(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const float alpha, const float* A, const nvpl_int_t lda, const nvpl_int_t stridea, const float* B, const nvpl_int_t ldb, const nvpl_int_t strideb, const float beta, float* C, const nvpl_int_t ldc, const nvpl_int_t stridec, const nvpl_int_t batch_size) except* nogil
|
|
164
|
+
cdef void cblas_dgemm_batch_strided(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const double alpha, const double* A, const nvpl_int_t lda, const nvpl_int_t stridea, const double* B, const nvpl_int_t ldb, const nvpl_int_t strideb, const double beta, double* C, const nvpl_int_t ldc, const nvpl_int_t stridec, const nvpl_int_t batch_size) except* nogil
|
|
165
|
+
cdef void cblas_cgemm_batch_strided(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const nvpl_int_t stridea, const void* B, const nvpl_int_t ldb, const nvpl_int_t strideb, const void* beta, void* C, const nvpl_int_t ldc, const nvpl_int_t stridec, const nvpl_int_t batch_size) except* nogil
|
|
166
|
+
cdef void cblas_zgemm_batch_strided(const CBLAS_ORDER Order, const CBLAS_TRANSPOSE TransA, const CBLAS_TRANSPOSE TransB, const nvpl_int_t M, const nvpl_int_t N, const nvpl_int_t K, const void* alpha, const void* A, const nvpl_int_t lda, const nvpl_int_t stridea, const void* B, const nvpl_int_t ldb, const nvpl_int_t strideb, const void* beta, void* C, const nvpl_int_t ldc, const nvpl_int_t stridec, const nvpl_int_t batch_size) except* nogil
|
|
Binary file
|
|
@@ -0,0 +1,92 @@
|
|
|
1
|
+
# Copyright (c) 2024-2026, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
|
|
2
|
+
#
|
|
3
|
+
# SPDX-License-Identifier: Apache-2.0
|
|
4
|
+
#
|
|
5
|
+
# This code was automatically generated with version 25.5, generator version 0.3.1.dev1303+g031f1197f. Do not modify it directly.
|
|
6
|
+
# This layer exposes the C header to Cython as-is.
|
|
7
|
+
|
|
8
|
+
###############################################################################
|
|
9
|
+
# Types (structs, enums, ...)
|
|
10
|
+
###############################################################################
|
|
11
|
+
|
|
12
|
+
# enums
|
|
13
|
+
|
|
14
|
+
|
|
15
|
+
|
|
16
|
+
# types
|
|
17
|
+
cdef extern from *:
|
|
18
|
+
"""
|
|
19
|
+
// Transform direction
|
|
20
|
+
#define FFTW_FORWARD -1
|
|
21
|
+
#define FFTW_INVERSE 1
|
|
22
|
+
#define FFTW_BACKWARD 1
|
|
23
|
+
|
|
24
|
+
// Planner flags
|
|
25
|
+
#define FFTW_ESTIMATE 0x01
|
|
26
|
+
#define FFTW_MEASURE 0x02
|
|
27
|
+
#define FFTW_PATIENT 0x03
|
|
28
|
+
#define FFTW_EXHAUSTIVE 0x04
|
|
29
|
+
#define FFTW_WISDOM_ONLY 0x05
|
|
30
|
+
|
|
31
|
+
// Algorithm restriction flags
|
|
32
|
+
#define FFTW_DESTROY_INPUT 0x08
|
|
33
|
+
#define FFTW_PRESERVE_INPUT 0x0C
|
|
34
|
+
#define FFTW_UNALIGNED 0x10
|
|
35
|
+
|
|
36
|
+
#ifdef _WIN32
|
|
37
|
+
typedef __declspec(align(16)) double fftw_complex[2];
|
|
38
|
+
typedef __declspec(align(8)) float fftwf_complex[2];
|
|
39
|
+
#else
|
|
40
|
+
typedef double fftw_complex[2] __attribute__ ((aligned (16)));
|
|
41
|
+
typedef float fftwf_complex[2] __attribute__ ((aligned (8)));
|
|
42
|
+
#endif
|
|
43
|
+
"""
|
|
44
|
+
|
|
45
|
+
cdef const int FFTW_FORWARD
|
|
46
|
+
cdef const int FFTW_INVERSE
|
|
47
|
+
|
|
48
|
+
cdef const int FFTW_ESTIMATE
|
|
49
|
+
cdef const int FFTW_MEASURE
|
|
50
|
+
cdef const int FFTW_PATIENT
|
|
51
|
+
cdef const int FFTW_EXHAUSTIVE
|
|
52
|
+
cdef const int FFTW_WISDOM_ONLY
|
|
53
|
+
|
|
54
|
+
cdef const int FFTW_DESTROY_INPUT
|
|
55
|
+
cdef const int FFTW_PRESERVE_INPUT
|
|
56
|
+
cdef const int FFTW_UNALIGNED
|
|
57
|
+
|
|
58
|
+
ctypedef double fftw_complex[2]
|
|
59
|
+
ctypedef float fftwf_complex[2]
|
|
60
|
+
|
|
61
|
+
|
|
62
|
+
ctypedef void* fftw_plan 'fftw_plan'
|
|
63
|
+
ctypedef void* fftwf_plan 'fftwf_plan'
|
|
64
|
+
|
|
65
|
+
|
|
66
|
+
###############################################################################
|
|
67
|
+
# Functions
|
|
68
|
+
###############################################################################
|
|
69
|
+
|
|
70
|
+
cdef int nvpl_fft_get_version() except?-42 nogil
|
|
71
|
+
cdef fftw_plan fftw_plan_many_dft(int rank, const int* n, int batch, fftw_complex* in_, const int* inembed, int istride, int idist, fftw_complex* out, const int* onembed, int ostride, int odist, int sign, unsigned flags) except?NULL nogil
|
|
72
|
+
cdef fftw_plan fftw_plan_many_dft_r2c(int rank, const int* n, int batch, double* in_, const int* inembed, int istride, int idist, fftw_complex* out, const int* onembed, int ostride, int odist, unsigned flags) except?NULL nogil
|
|
73
|
+
cdef fftw_plan fftw_plan_many_dft_c2r(int rank, const int* n, int batch, fftw_complex* in_, const int* inembed, int istride, int idist, double* out, const int* onembed, int ostride, int odist, unsigned flags) except?NULL nogil
|
|
74
|
+
cdef void fftw_execute_dft(const fftw_plan plan, fftw_complex* idata, fftw_complex* odata) except* nogil
|
|
75
|
+
cdef void fftw_execute_dft_r2c(const fftw_plan plan, double* idata, fftw_complex* odata) except* nogil
|
|
76
|
+
cdef void fftw_execute_dft_c2r(const fftw_plan plan, fftw_complex* idata, double* odata) except* nogil
|
|
77
|
+
cdef fftwf_plan fftwf_plan_many_dft(int rank, const int* n, int batch, fftwf_complex* in_, const int* inembed, int istride, int idist, fftwf_complex* out, const int* onembed, int ostride, int odist, int sign, unsigned flags) except?NULL nogil
|
|
78
|
+
cdef fftwf_plan fftwf_plan_many_dft_r2c(int rank, const int* n, int batch, float* in_, const int* inembed, int istride, int idist, fftwf_complex* out, const int* onembed, int ostride, int odist, unsigned flags) except?NULL nogil
|
|
79
|
+
cdef fftwf_plan fftwf_plan_many_dft_c2r(int rank, const int* n, int batch, fftwf_complex* in_, const int* inembed, int istride, int idist, float* out, const int* onembed, int ostride, int odist, unsigned flags) except?NULL nogil
|
|
80
|
+
cdef void fftwf_execute_dft(const fftwf_plan plan, fftwf_complex* idata, fftwf_complex* odata) except* nogil
|
|
81
|
+
cdef void fftwf_execute_dft_r2c(const fftwf_plan plan, float* idata, fftwf_complex* odata) except* nogil
|
|
82
|
+
cdef void fftwf_execute_dft_c2r(const fftwf_plan plan, fftwf_complex* idata, float* odata) except* nogil
|
|
83
|
+
cdef int fftw_init_threads() except?-42 nogil
|
|
84
|
+
cdef int fftwf_init_threads() except?-42 nogil
|
|
85
|
+
cdef void fftw_plan_with_nthreads(int nthreads) except* nogil
|
|
86
|
+
cdef void fftwf_plan_with_nthreads(int nthreads) except* nogil
|
|
87
|
+
cdef int fftw_planner_nthreads() except?-42 nogil
|
|
88
|
+
cdef int fftwf_planner_nthreads() except?-42 nogil
|
|
89
|
+
cdef void fftw_cleanup_threads() except* nogil
|
|
90
|
+
cdef void fftwf_cleanup_threads() except* nogil
|
|
91
|
+
cdef void fftw_destroy_plan(fftw_plan plan) except* nogil
|
|
92
|
+
cdef void fftwf_destroy_plan(fftwf_plan plan) except* nogil
|
|
Binary file
|
|
@@ -0,0 +1,100 @@
|
|
|
1
|
+
# Copyright (c) 2024-2026, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
|
|
2
|
+
#
|
|
3
|
+
# SPDX-License-Identifier: Apache-2.0
|
|
4
|
+
#
|
|
5
|
+
# This code was automatically generated with version 25.5, generator version 0.3.1.dev1303+g031f1197f. Do not modify it directly.
|
|
6
|
+
|
|
7
|
+
from libc.stdint cimport intptr_t
|
|
8
|
+
|
|
9
|
+
from .cyfft cimport *
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
###############################################################################
|
|
13
|
+
# Types
|
|
14
|
+
###############################################################################
|
|
15
|
+
|
|
16
|
+
|
|
17
|
+
|
|
18
|
+
|
|
19
|
+
cdef class Plan:
|
|
20
|
+
cdef intptr_t forward_handle
|
|
21
|
+
cdef intptr_t inverse_handle
|
|
22
|
+
cdef int precision
|
|
23
|
+
cdef int kind
|
|
24
|
+
|
|
25
|
+
|
|
26
|
+
###############################################################################
|
|
27
|
+
# Enum
|
|
28
|
+
###############################################################################
|
|
29
|
+
|
|
30
|
+
|
|
31
|
+
# The `UNSPECIFIED` value is not valid sign in NVPL's header,
|
|
32
|
+
# but is used here to allow for planning C2C that can handle
|
|
33
|
+
# both directions
|
|
34
|
+
cpdef enum Sign:
|
|
35
|
+
UNSPECIFIED = 0
|
|
36
|
+
FORWARD = FFTW_FORWARD
|
|
37
|
+
INVERSE = FFTW_INVERSE
|
|
38
|
+
|
|
39
|
+
|
|
40
|
+
cpdef enum PlannerFlags:
|
|
41
|
+
ESTIMATE = FFTW_ESTIMATE
|
|
42
|
+
MEASURE = FFTW_MEASURE
|
|
43
|
+
PATIENT = FFTW_PATIENT
|
|
44
|
+
EXHAUSTIVE = FFTW_EXHAUSTIVE
|
|
45
|
+
WISDOM_ONLY = FFTW_WISDOM_ONLY
|
|
46
|
+
|
|
47
|
+
|
|
48
|
+
cpdef enum Precision:
|
|
49
|
+
FLOAT = 32
|
|
50
|
+
DOUBLE = 64
|
|
51
|
+
|
|
52
|
+
|
|
53
|
+
cpdef enum Kind:
|
|
54
|
+
C2C = 1
|
|
55
|
+
C2R = 2
|
|
56
|
+
R2C = 3
|
|
57
|
+
|
|
58
|
+
|
|
59
|
+
|
|
60
|
+
|
|
61
|
+
|
|
62
|
+
###############################################################################
|
|
63
|
+
# Convenience wrappers/adapters
|
|
64
|
+
###############################################################################
|
|
65
|
+
|
|
66
|
+
cpdef int plan_with_nthreads(int precision, int num_threads) except -1
|
|
67
|
+
cpdef int planner_nthreads(int precision) except? -1
|
|
68
|
+
cpdef int init_threads(int precision) except -1
|
|
69
|
+
cpdef int cleanup_threads(int precision) except -1
|
|
70
|
+
cpdef Plan plan_many(int precision, int kind, int sign, int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, unsigned flags)
|
|
71
|
+
cpdef int destroy(Plan plan) except -1
|
|
72
|
+
cpdef int execute(Plan plan, intptr_t idata, intptr_t odata, int sign) except -1
|
|
73
|
+
|
|
74
|
+
###############################################################################
|
|
75
|
+
# Functions
|
|
76
|
+
###############################################################################
|
|
77
|
+
|
|
78
|
+
cpdef int get_version() except? -1
|
|
79
|
+
cpdef intptr_t plan_many_c2c_double(int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, int sign, unsigned flags) except? 0
|
|
80
|
+
cpdef intptr_t plan_many_r2c_double(int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, unsigned flags) except? 0
|
|
81
|
+
cpdef intptr_t plan_many_c2r_double(int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, unsigned flags) except? 0
|
|
82
|
+
cpdef void execute_c2c_double(intptr_t plan, intptr_t idata, intptr_t odata) except*
|
|
83
|
+
cpdef void execute_r2c_double(intptr_t plan, intptr_t idata, intptr_t odata) except*
|
|
84
|
+
cpdef void execute_c2r_double(intptr_t plan, intptr_t idata, intptr_t odata) except*
|
|
85
|
+
cpdef intptr_t plan_many_c2c_float(int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, int sign, unsigned flags) except? 0
|
|
86
|
+
cpdef intptr_t plan_many_r2c_float(int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, unsigned flags) except? 0
|
|
87
|
+
cpdef intptr_t plan_many_c2r_float(int rank, n, int batch, intptr_t in_, inembed, int istride, int idist, intptr_t out, onembed, int ostride, int odist, unsigned flags) except? 0
|
|
88
|
+
cpdef void execute_c2c_float(intptr_t plan, intptr_t idata, intptr_t odata) except*
|
|
89
|
+
cpdef void execute_r2c_float(intptr_t plan, intptr_t idata, intptr_t odata) except*
|
|
90
|
+
cpdef void execute_c2r_float(intptr_t plan, intptr_t idata, intptr_t odata) except*
|
|
91
|
+
cpdef int init_threads_double() except 0
|
|
92
|
+
cpdef int init_threads_float() except 0
|
|
93
|
+
cpdef void plan_with_nthreads_double(int nthreads) except*
|
|
94
|
+
cpdef void plan_with_nthreads_float(int nthreads) except*
|
|
95
|
+
cpdef int planner_nthreads_double() except? 0
|
|
96
|
+
cpdef int planner_nthreads_float() except? 0
|
|
97
|
+
cpdef void cleanup_threads_double() except*
|
|
98
|
+
cpdef void cleanup_threads_float() except*
|
|
99
|
+
cpdef void destroy_plan_double(intptr_t plan) except*
|
|
100
|
+
cpdef void destroy_plan_float(intptr_t plan) except*
|
|
@@ -0,0 +1,100 @@
|
|
|
1
|
+
# Copyright (c) 2026, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
|
|
2
|
+
#
|
|
3
|
+
# SPDX-License-Identifier: Apache-2.0
|
|
4
|
+
|
|
5
|
+
import _cython_3_2_4
|
|
6
|
+
import enum
|
|
7
|
+
from typing import Any, Callable, ClassVar
|
|
8
|
+
|
|
9
|
+
__pyx_capi__: dict
|
|
10
|
+
__test__: dict
|
|
11
|
+
cleanup_threads: _cython_3_2_4.cython_function_or_method
|
|
12
|
+
cleanup_threads_double: _cython_3_2_4.cython_function_or_method
|
|
13
|
+
cleanup_threads_float: _cython_3_2_4.cython_function_or_method
|
|
14
|
+
destroy: _cython_3_2_4.cython_function_or_method
|
|
15
|
+
destroy_plan_double: _cython_3_2_4.cython_function_or_method
|
|
16
|
+
destroy_plan_float: _cython_3_2_4.cython_function_or_method
|
|
17
|
+
execute: _cython_3_2_4.cython_function_or_method
|
|
18
|
+
execute_c2c_double: _cython_3_2_4.cython_function_or_method
|
|
19
|
+
execute_c2c_float: _cython_3_2_4.cython_function_or_method
|
|
20
|
+
execute_c2r_double: _cython_3_2_4.cython_function_or_method
|
|
21
|
+
execute_c2r_float: _cython_3_2_4.cython_function_or_method
|
|
22
|
+
execute_r2c_double: _cython_3_2_4.cython_function_or_method
|
|
23
|
+
execute_r2c_float: _cython_3_2_4.cython_function_or_method
|
|
24
|
+
get_version: _cython_3_2_4.cython_function_or_method
|
|
25
|
+
init_threads: _cython_3_2_4.cython_function_or_method
|
|
26
|
+
init_threads_double: _cython_3_2_4.cython_function_or_method
|
|
27
|
+
init_threads_float: _cython_3_2_4.cython_function_or_method
|
|
28
|
+
plan_many: _cython_3_2_4.cython_function_or_method
|
|
29
|
+
plan_many_c2c_double: _cython_3_2_4.cython_function_or_method
|
|
30
|
+
plan_many_c2c_float: _cython_3_2_4.cython_function_or_method
|
|
31
|
+
plan_many_c2r_double: _cython_3_2_4.cython_function_or_method
|
|
32
|
+
plan_many_c2r_float: _cython_3_2_4.cython_function_or_method
|
|
33
|
+
plan_many_r2c_double: _cython_3_2_4.cython_function_or_method
|
|
34
|
+
plan_many_r2c_float: _cython_3_2_4.cython_function_or_method
|
|
35
|
+
plan_with_nthreads: _cython_3_2_4.cython_function_or_method
|
|
36
|
+
plan_with_nthreads_double: _cython_3_2_4.cython_function_or_method
|
|
37
|
+
plan_with_nthreads_float: _cython_3_2_4.cython_function_or_method
|
|
38
|
+
planner_nthreads: _cython_3_2_4.cython_function_or_method
|
|
39
|
+
planner_nthreads_double: _cython_3_2_4.cython_function_or_method
|
|
40
|
+
planner_nthreads_float: _cython_3_2_4.cython_function_or_method
|
|
41
|
+
|
|
42
|
+
class FFTWError(Exception): ...
|
|
43
|
+
|
|
44
|
+
class FFTWUnaligned(FFTWError): ...
|
|
45
|
+
|
|
46
|
+
class Kind(enum.IntFlag):
|
|
47
|
+
"""An enumeration."""
|
|
48
|
+
__new__: ClassVar[Callable] = ...
|
|
49
|
+
C2C: ClassVar[Kind] = ...
|
|
50
|
+
C2R: ClassVar[Kind] = ...
|
|
51
|
+
R2C: ClassVar[Kind] = ...
|
|
52
|
+
_generate_next_value_: ClassVar[Callable] = ...
|
|
53
|
+
_member_map_: ClassVar[dict] = ...
|
|
54
|
+
_member_names_: ClassVar[list] = ...
|
|
55
|
+
_member_type_: ClassVar[type[int]] = ...
|
|
56
|
+
_value2member_map_: ClassVar[dict] = ...
|
|
57
|
+
|
|
58
|
+
class Plan:
|
|
59
|
+
@classmethod
|
|
60
|
+
def __init__(cls, *args, **kwargs) -> None:
|
|
61
|
+
"""Create and return a new object. See help(type) for accurate signature."""
|
|
62
|
+
def __reduce__(self) -> Any:
|
|
63
|
+
"""Plan.__reduce__(self)"""
|
|
64
|
+
|
|
65
|
+
class PlannerFlags(enum.IntFlag):
|
|
66
|
+
"""An enumeration."""
|
|
67
|
+
__new__: ClassVar[Callable] = ...
|
|
68
|
+
ESTIMATE: ClassVar[PlannerFlags] = ...
|
|
69
|
+
EXHAUSTIVE: ClassVar[PlannerFlags] = ...
|
|
70
|
+
MEASURE: ClassVar[PlannerFlags] = ...
|
|
71
|
+
PATIENT: ClassVar[PlannerFlags] = ...
|
|
72
|
+
WISDOM_ONLY: ClassVar[PlannerFlags] = ...
|
|
73
|
+
_generate_next_value_: ClassVar[Callable] = ...
|
|
74
|
+
_member_map_: ClassVar[dict] = ...
|
|
75
|
+
_member_names_: ClassVar[list] = ...
|
|
76
|
+
_member_type_: ClassVar[type[int]] = ...
|
|
77
|
+
_value2member_map_: ClassVar[dict] = ...
|
|
78
|
+
|
|
79
|
+
class Precision(enum.IntFlag):
|
|
80
|
+
"""An enumeration."""
|
|
81
|
+
__new__: ClassVar[Callable] = ...
|
|
82
|
+
DOUBLE: ClassVar[Precision] = ...
|
|
83
|
+
FLOAT: ClassVar[Precision] = ...
|
|
84
|
+
_generate_next_value_: ClassVar[Callable] = ...
|
|
85
|
+
_member_map_: ClassVar[dict] = ...
|
|
86
|
+
_member_names_: ClassVar[list] = ...
|
|
87
|
+
_member_type_: ClassVar[type[int]] = ...
|
|
88
|
+
_value2member_map_: ClassVar[dict] = ...
|
|
89
|
+
|
|
90
|
+
class Sign(enum.IntFlag):
|
|
91
|
+
"""An enumeration."""
|
|
92
|
+
__new__: ClassVar[Callable] = ...
|
|
93
|
+
FORWARD: ClassVar[Sign] = ...
|
|
94
|
+
INVERSE: ClassVar[Sign] = ...
|
|
95
|
+
UNSPECIFIED: ClassVar[Sign] = ...
|
|
96
|
+
_generate_next_value_: ClassVar[Callable] = ...
|
|
97
|
+
_member_map_: ClassVar[dict] = ...
|
|
98
|
+
_member_names_: ClassVar[list] = ...
|
|
99
|
+
_member_type_: ClassVar[type[int]] = ...
|
|
100
|
+
_value2member_map_: ClassVar[dict] = ...
|
|
@@ -0,0 +1,54 @@
|
|
|
1
|
+
# Copyright (c) 2025-2026, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
|
|
2
|
+
#
|
|
3
|
+
# SPDX-License-Identifier: Apache-2.0
|
|
4
|
+
#
|
|
5
|
+
# This code was automatically generated with version 3.1.7. Do not modify it directly.
|
|
6
|
+
|
|
7
|
+
from libc.stdint cimport intptr_t
|
|
8
|
+
|
|
9
|
+
from .cynvshmem cimport *
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
###############################################################################
|
|
13
|
+
# Types
|
|
14
|
+
###############################################################################
|
|
15
|
+
|
|
16
|
+
ctypedef nvshmem_team_config_v1 team_config
|
|
17
|
+
ctypedef nvshmemx_uniqueid_args_v1 uniqueid_args
|
|
18
|
+
ctypedef nvshmemx_init_args_v1 init_args
|
|
19
|
+
ctypedef nvshmemx_init_attr_v1 init_attr
|
|
20
|
+
|
|
21
|
+
ctypedef cudaStream_t Stream
|
|
22
|
+
|
|
23
|
+
|
|
24
|
+
###############################################################################
|
|
25
|
+
# Enum
|
|
26
|
+
###############################################################################
|
|
27
|
+
|
|
28
|
+
ctypedef _anon_enum0 __anon_enum0
|
|
29
|
+
ctypedef _anon_enum1 __anon_enum1
|
|
30
|
+
ctypedef _anon_enum2 __anon_enum2
|
|
31
|
+
ctypedef nvshmemx_status _Status
|
|
32
|
+
ctypedef flags _Flags
|
|
33
|
+
|
|
34
|
+
|
|
35
|
+
###############################################################################
|
|
36
|
+
# Functions
|
|
37
|
+
###############################################################################
|
|
38
|
+
|
|
39
|
+
cpdef int init_status() except? -1
|
|
40
|
+
cpdef int my_pe() except? -1
|
|
41
|
+
cpdef int n_pes() except? -1
|
|
42
|
+
cpdef intptr_t malloc(size_t size) except? 0
|
|
43
|
+
cpdef intptr_t calloc(size_t count, size_t size) except? 0
|
|
44
|
+
cpdef intptr_t align(size_t alignment, size_t size) except? 0
|
|
45
|
+
cpdef void free(intptr_t ptr) except*
|
|
46
|
+
cpdef intptr_t ptr(intptr_t dest, int pe) except? 0
|
|
47
|
+
cpdef void int_p(intptr_t dest, int value, int pe) except*
|
|
48
|
+
cpdef int team_my_pe(int32_t team) except? -1
|
|
49
|
+
cpdef void barrier_all_on_stream(intptr_t stream) except*
|
|
50
|
+
cpdef void sync_all_on_stream(intptr_t stream) except*
|
|
51
|
+
cpdef hostlib_init_attr(unsigned int flags, intptr_t attr)
|
|
52
|
+
cpdef void hostlib_finalize() except*
|
|
53
|
+
cpdef set_attr_uniqueid_args(int myrank, int nranks, intptr_t uniqueid, intptr_t attr)
|
|
54
|
+
cpdef get_uniqueid(intptr_t uniqueid)
|