Add Chromium-only Blender WebEngine parity work

blender-5.2.0/extern/cuew/CMakeLists.txt

@@ -0,0 +1,23 @@
+# SPDX-FileCopyrightText: 2006 Blender Foundation
+#
+# SPDX-License-Identifier: GPL-2.0-or-later
+
+set(INC
+  .
+  include
+)
+
+set(INC_SYS
+
+)
+
+set(SRC
+  src/cuew.c
+
+  include/cuew.h
+)
+
+set(LIB
+)
+
+blender_add_lib(extern_cuew "${SRC}" "${INC}" "${INC_SYS}" "${LIB}")

blender-5.2.0/extern/cuew/LICENSE

@@ -0,0 +1,174 @@
+
+                               Modified Apache 2.0 License
+
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor
+      and its affiliates, except as required to comply with Section 4(c) of
+      the License and to reproduce the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.

blender-5.2.0/extern/cuew/README

@@ -0,0 +1,12 @@
+The CUDA Extension Wrangler Library (CUEW) is a cross-platform open-source
+C/C++ extension loading library. CUEW provides efficient run-time mechanisms
+for determining which CUDA functions and extensions extensions are supported
+on the target platform.
+
+CUDA core and extension functionality is exposed in a single header file.
+CUEW has been tested on a variety of operating systems, including Windows,
+Linux, Mac OS X.
+
+LICENSE
+
+CUEW library is released under the Apache 2.0 license.

blender-5.2.0/extern/cuew/README.blender

@@ -0,0 +1,6 @@
+Project: Cuda Wrangler
+URL: https://github.com/CudaWrangler/cuew
+License: SPDX:Apache-2.0
+Upstream version: cbf465b
+Copyright: Copyright 2011-2014 Blender Foundation
+Local modifications: None

blender-5.2.0/extern/cuew/src/cuew.c

@@ -0,0 +1,926 @@
+/*
+ * Copyright 2011-2014 Blender Foundation
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License
+ */
+
+#ifdef _MSC_VER
+#  if _MSC_VER < 1900
+#    define snprintf _snprintf
+#  endif
+#  define popen _popen
+#  define pclose _pclose
+#  define _CRT_SECURE_NO_WARNINGS
+#endif
+
+#include <cuew.h>
+#include <assert.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/stat.h>
+
+#ifdef _WIN32
+#  ifndef WIN32_LEAN_AND_MEAN
+#    define WIN32_LEAN_AND_MEAN
+#  endif
+#  define VC_EXTRALEAN
+#  include <windows.h>
+
+/* Utility macros. */
+
+typedef HMODULE DynamicLibrary;
+
+#  define dynamic_library_open(path)         LoadLibraryA(path)
+#  define dynamic_library_close(lib)         FreeLibrary(lib)
+#  define dynamic_library_find(lib, symbol)  GetProcAddress(lib, symbol)
+#else
+#  include <dlfcn.h>
+
+typedef void* DynamicLibrary;
+
+#  define dynamic_library_open(path)         dlopen(path, RTLD_NOW)
+#  define dynamic_library_close(lib)         dlclose(lib)
+#  define dynamic_library_find(lib, symbol)  dlsym(lib, symbol)
+#endif
+
+#define _LIBRARY_FIND_CHECKED(lib, name) \
+        name = (t##name *)dynamic_library_find(lib, #name); \
+        assert(name);
+
+#define _LIBRARY_FIND(lib, name) \
+        name = (t##name *)dynamic_library_find(lib, #name);
+
+#define CUDA_LIBRARY_FIND_CHECKED(name) \
+        _LIBRARY_FIND_CHECKED(cuda_lib, name)
+#define CUDA_LIBRARY_FIND(name) _LIBRARY_FIND(cuda_lib, name)
+
+#define NVRTC_LIBRARY_FIND_CHECKED(name) \
+        _LIBRARY_FIND_CHECKED(nvrtc_lib, name)
+#define NVRTC_LIBRARY_FIND(name) _LIBRARY_FIND(nvrtc_lib, name)
+
+static DynamicLibrary cuda_lib;
+static DynamicLibrary nvrtc_lib;
+
+/* Function definitions. */
+tcuGetErrorString *cuGetErrorString;
+tcuGetErrorName *cuGetErrorName;
+tcuInit *cuInit;
+tcuDriverGetVersion *cuDriverGetVersion;
+tcuDeviceGet *cuDeviceGet;
+tcuDeviceGetCount *cuDeviceGetCount;
+tcuDeviceGetName *cuDeviceGetName;
+tcuDeviceGetUuid *cuDeviceGetUuid;
+tcuDeviceTotalMem_v2 *cuDeviceTotalMem_v2;
+tcuDeviceGetAttribute *cuDeviceGetAttribute;
+tcuDeviceGetProperties *cuDeviceGetProperties;
+tcuDeviceComputeCapability *cuDeviceComputeCapability;
+tcuDevicePrimaryCtxRetain *cuDevicePrimaryCtxRetain;
+tcuDevicePrimaryCtxRelease *cuDevicePrimaryCtxRelease;
+tcuDevicePrimaryCtxSetFlags *cuDevicePrimaryCtxSetFlags;
+tcuDevicePrimaryCtxGetState *cuDevicePrimaryCtxGetState;
+tcuDevicePrimaryCtxReset *cuDevicePrimaryCtxReset;
+tcuCtxCreate_v2 *cuCtxCreate_v2;
+tcuCtxDestroy_v2 *cuCtxDestroy_v2;
+tcuCtxPushCurrent_v2 *cuCtxPushCurrent_v2;
+tcuCtxPopCurrent_v2 *cuCtxPopCurrent_v2;
+tcuCtxSetCurrent *cuCtxSetCurrent;
+tcuCtxGetCurrent *cuCtxGetCurrent;
+tcuCtxGetDevice *cuCtxGetDevice;
+tcuCtxGetFlags *cuCtxGetFlags;
+tcuCtxSynchronize *cuCtxSynchronize;
+tcuCtxSetLimit *cuCtxSetLimit;
+tcuCtxGetLimit *cuCtxGetLimit;
+tcuCtxGetCacheConfig *cuCtxGetCacheConfig;
+tcuCtxSetCacheConfig *cuCtxSetCacheConfig;
+tcuCtxGetSharedMemConfig *cuCtxGetSharedMemConfig;
+tcuCtxSetSharedMemConfig *cuCtxSetSharedMemConfig;
+tcuCtxGetApiVersion *cuCtxGetApiVersion;
+tcuCtxGetStreamPriorityRange *cuCtxGetStreamPriorityRange;
+tcuCtxAttach *cuCtxAttach;
+tcuCtxDetach *cuCtxDetach;
+tcuModuleLoad *cuModuleLoad;
+tcuModuleLoadData *cuModuleLoadData;
+tcuModuleLoadDataEx *cuModuleLoadDataEx;
+tcuModuleLoadFatBinary *cuModuleLoadFatBinary;
+tcuModuleUnload *cuModuleUnload;
+tcuModuleGetFunction *cuModuleGetFunction;
+tcuModuleGetGlobal_v2 *cuModuleGetGlobal_v2;
+tcuModuleGetTexRef *cuModuleGetTexRef;
+tcuModuleGetSurfRef *cuModuleGetSurfRef;
+tcuLinkCreate_v2 *cuLinkCreate_v2;
+tcuLinkAddData_v2 *cuLinkAddData_v2;
+tcuLinkAddFile_v2 *cuLinkAddFile_v2;
+tcuLinkComplete *cuLinkComplete;
+tcuLinkDestroy *cuLinkDestroy;
+tcuMemGetInfo_v2 *cuMemGetInfo_v2;
+tcuMemAlloc_v2 *cuMemAlloc_v2;
+tcuMemAllocPitch_v2 *cuMemAllocPitch_v2;
+tcuMemFree_v2 *cuMemFree_v2;
+tcuMemGetAddressRange_v2 *cuMemGetAddressRange_v2;
+tcuMemAllocHost_v2 *cuMemAllocHost_v2;
+tcuMemFreeHost *cuMemFreeHost;
+tcuMemHostAlloc *cuMemHostAlloc;
+tcuMemHostGetDevicePointer_v2 *cuMemHostGetDevicePointer_v2;
+tcuMemHostGetFlags *cuMemHostGetFlags;
+tcuMemAllocManaged *cuMemAllocManaged;
+tcuDeviceGetByPCIBusId *cuDeviceGetByPCIBusId;
+tcuDeviceGetPCIBusId *cuDeviceGetPCIBusId;
+tcuIpcGetEventHandle *cuIpcGetEventHandle;
+tcuIpcOpenEventHandle *cuIpcOpenEventHandle;
+tcuIpcGetMemHandle *cuIpcGetMemHandle;
+tcuIpcOpenMemHandle *cuIpcOpenMemHandle;
+tcuIpcCloseMemHandle *cuIpcCloseMemHandle;
+tcuMemHostRegister_v2 *cuMemHostRegister_v2;
+tcuMemHostUnregister *cuMemHostUnregister;
+tcuMemcpy *cuMemcpy;
+tcuMemcpyPeer *cuMemcpyPeer;
+tcuMemcpyHtoD_v2 *cuMemcpyHtoD_v2;
+tcuMemcpyDtoH_v2 *cuMemcpyDtoH_v2;
+tcuMemcpyDtoD_v2 *cuMemcpyDtoD_v2;
+tcuMemcpyDtoA_v2 *cuMemcpyDtoA_v2;
+tcuMemcpyAtoD_v2 *cuMemcpyAtoD_v2;
+tcuMemcpyHtoA_v2 *cuMemcpyHtoA_v2;
+tcuMemcpyAtoH_v2 *cuMemcpyAtoH_v2;
+tcuMemcpyAtoA_v2 *cuMemcpyAtoA_v2;
+tcuMemcpy2D_v2 *cuMemcpy2D_v2;
+tcuMemcpy2DUnaligned_v2 *cuMemcpy2DUnaligned_v2;
+tcuMemcpy3D_v2 *cuMemcpy3D_v2;
+tcuMemcpy3DPeer *cuMemcpy3DPeer;
+tcuMemcpyAsync *cuMemcpyAsync;
+tcuMemcpyPeerAsync *cuMemcpyPeerAsync;
+tcuMemcpyHtoDAsync_v2 *cuMemcpyHtoDAsync_v2;
+tcuMemcpyDtoHAsync_v2 *cuMemcpyDtoHAsync_v2;
+tcuMemcpyDtoDAsync_v2 *cuMemcpyDtoDAsync_v2;
+tcuMemcpyHtoAAsync_v2 *cuMemcpyHtoAAsync_v2;
+tcuMemcpyAtoHAsync_v2 *cuMemcpyAtoHAsync_v2;
+tcuMemcpy2DAsync_v2 *cuMemcpy2DAsync_v2;
+tcuMemcpy3DAsync_v2 *cuMemcpy3DAsync_v2;
+tcuMemcpy3DPeerAsync *cuMemcpy3DPeerAsync;
+tcuMemsetD8_v2 *cuMemsetD8_v2;
+tcuMemsetD16_v2 *cuMemsetD16_v2;
+tcuMemsetD32_v2 *cuMemsetD32_v2;
+tcuMemsetD2D8_v2 *cuMemsetD2D8_v2;
+tcuMemsetD2D16_v2 *cuMemsetD2D16_v2;
+tcuMemsetD2D32_v2 *cuMemsetD2D32_v2;
+tcuMemsetD8Async *cuMemsetD8Async;
+tcuMemsetD16Async *cuMemsetD16Async;
+tcuMemsetD32Async *cuMemsetD32Async;
+tcuMemsetD2D8Async *cuMemsetD2D8Async;
+tcuMemsetD2D16Async *cuMemsetD2D16Async;
+tcuMemsetD2D32Async *cuMemsetD2D32Async;
+tcuArrayCreate_v2 *cuArrayCreate_v2;
+tcuArrayGetDescriptor_v2 *cuArrayGetDescriptor_v2;
+tcuArrayDestroy *cuArrayDestroy;
+tcuArray3DCreate_v2 *cuArray3DCreate_v2;
+tcuArray3DGetDescriptor_v2 *cuArray3DGetDescriptor_v2;
+tcuMipmappedArrayCreate *cuMipmappedArrayCreate;
+tcuMipmappedArrayGetLevel *cuMipmappedArrayGetLevel;
+tcuMipmappedArrayDestroy *cuMipmappedArrayDestroy;
+tcuPointerGetAttribute *cuPointerGetAttribute;
+tcuMemPrefetchAsync *cuMemPrefetchAsync;
+tcuMemAdvise *cuMemAdvise;
+tcuMemRangeGetAttribute *cuMemRangeGetAttribute;
+tcuMemRangeGetAttributes *cuMemRangeGetAttributes;
+tcuPointerSetAttribute *cuPointerSetAttribute;
+tcuPointerGetAttributes *cuPointerGetAttributes;
+tcuStreamCreate *cuStreamCreate;
+tcuStreamCreateWithPriority *cuStreamCreateWithPriority;
+tcuStreamGetPriority *cuStreamGetPriority;
+tcuStreamGetFlags *cuStreamGetFlags;
+tcuStreamGetCtx *cuStreamGetCtx;
+tcuStreamWaitEvent *cuStreamWaitEvent;
+tcuStreamAddCallback *cuStreamAddCallback;
+tcuStreamAttachMemAsync *cuStreamAttachMemAsync;
+tcuStreamQuery *cuStreamQuery;
+tcuStreamSynchronize *cuStreamSynchronize;
+tcuStreamDestroy_v2 *cuStreamDestroy_v2;
+tcuEventCreate *cuEventCreate;
+tcuEventRecord *cuEventRecord;
+tcuEventQuery *cuEventQuery;
+tcuEventSynchronize *cuEventSynchronize;
+tcuEventDestroy_v2 *cuEventDestroy_v2;
+tcuEventElapsedTime *cuEventElapsedTime;
+tcuStreamWaitValue32 *cuStreamWaitValue32;
+tcuStreamWaitValue64 *cuStreamWaitValue64;
+tcuStreamWriteValue32 *cuStreamWriteValue32;
+tcuStreamWriteValue64 *cuStreamWriteValue64;
+tcuStreamBatchMemOp *cuStreamBatchMemOp;
+tcuFuncGetAttribute *cuFuncGetAttribute;
+tcuFuncSetAttribute *cuFuncSetAttribute;
+tcuFuncSetCacheConfig *cuFuncSetCacheConfig;
+tcuFuncSetSharedMemConfig *cuFuncSetSharedMemConfig;
+tcuLaunchKernel *cuLaunchKernel;
+tcuLaunchCooperativeKernel *cuLaunchCooperativeKernel;
+tcuLaunchCooperativeKernelMultiDevice *cuLaunchCooperativeKernelMultiDevice;
+tcuFuncSetBlockShape *cuFuncSetBlockShape;
+tcuFuncSetSharedSize *cuFuncSetSharedSize;
+tcuParamSetSize *cuParamSetSize;
+tcuParamSeti *cuParamSeti;
+tcuParamSetf *cuParamSetf;
+tcuParamSetv *cuParamSetv;
+tcuLaunch *cuLaunch;
+tcuLaunchGrid *cuLaunchGrid;
+tcuLaunchGridAsync *cuLaunchGridAsync;
+tcuParamSetTexRef *cuParamSetTexRef;
+tcuOccupancyMaxActiveBlocksPerMultiprocessor *cuOccupancyMaxActiveBlocksPerMultiprocessor;
+tcuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags *cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags;
+tcuOccupancyMaxPotentialBlockSize *cuOccupancyMaxPotentialBlockSize;
+tcuOccupancyMaxPotentialBlockSizeWithFlags *cuOccupancyMaxPotentialBlockSizeWithFlags;
+tcuTexRefSetArray *cuTexRefSetArray;
+tcuTexRefSetMipmappedArray *cuTexRefSetMipmappedArray;
+tcuTexRefSetAddress_v2 *cuTexRefSetAddress_v2;
+tcuTexRefSetAddress2D_v3 *cuTexRefSetAddress2D_v3;
+tcuTexRefSetFormat *cuTexRefSetFormat;
+tcuTexRefSetAddressMode *cuTexRefSetAddressMode;
+tcuTexRefSetFilterMode *cuTexRefSetFilterMode;
+tcuTexRefSetMipmapFilterMode *cuTexRefSetMipmapFilterMode;
+tcuTexRefSetMipmapLevelBias *cuTexRefSetMipmapLevelBias;
+tcuTexRefSetMipmapLevelClamp *cuTexRefSetMipmapLevelClamp;
+tcuTexRefSetMaxAnisotropy *cuTexRefSetMaxAnisotropy;
+tcuTexRefSetBorderColor *cuTexRefSetBorderColor;
+tcuTexRefSetFlags *cuTexRefSetFlags;
+tcuTexRefGetAddress_v2 *cuTexRefGetAddress_v2;
+tcuTexRefGetArray *cuTexRefGetArray;
+tcuTexRefGetMipmappedArray *cuTexRefGetMipmappedArray;
+tcuTexRefGetAddressMode *cuTexRefGetAddressMode;
+tcuTexRefGetFilterMode *cuTexRefGetFilterMode;
+tcuTexRefGetFormat *cuTexRefGetFormat;
+tcuTexRefGetMipmapFilterMode *cuTexRefGetMipmapFilterMode;
+tcuTexRefGetMipmapLevelBias *cuTexRefGetMipmapLevelBias;
+tcuTexRefGetMipmapLevelClamp *cuTexRefGetMipmapLevelClamp;
+tcuTexRefGetMaxAnisotropy *cuTexRefGetMaxAnisotropy;
+tcuTexRefGetBorderColor *cuTexRefGetBorderColor;
+tcuTexRefGetFlags *cuTexRefGetFlags;
+tcuTexRefCreate *cuTexRefCreate;
+tcuTexRefDestroy *cuTexRefDestroy;
+tcuSurfRefSetArray *cuSurfRefSetArray;
+tcuSurfRefGetArray *cuSurfRefGetArray;
+tcuTexObjectCreate *cuTexObjectCreate;
+tcuTexObjectDestroy *cuTexObjectDestroy;
+tcuTexObjectGetResourceDesc *cuTexObjectGetResourceDesc;
+tcuTexObjectGetTextureDesc *cuTexObjectGetTextureDesc;
+tcuTexObjectGetResourceViewDesc *cuTexObjectGetResourceViewDesc;
+tcuSurfObjectCreate *cuSurfObjectCreate;
+tcuSurfObjectDestroy *cuSurfObjectDestroy;
+tcuSurfObjectGetResourceDesc *cuSurfObjectGetResourceDesc;
+tcuDeviceCanAccessPeer *cuDeviceCanAccessPeer;
+tcuCtxEnablePeerAccess *cuCtxEnablePeerAccess;
+tcuCtxDisablePeerAccess *cuCtxDisablePeerAccess;
+tcuDeviceGetP2PAttribute *cuDeviceGetP2PAttribute;
+tcuGraphicsUnregisterResource *cuGraphicsUnregisterResource;
+tcuGraphicsSubResourceGetMappedArray *cuGraphicsSubResourceGetMappedArray;
+tcuGraphicsResourceGetMappedMipmappedArray *cuGraphicsResourceGetMappedMipmappedArray;
+tcuGraphicsResourceGetMappedPointer_v2 *cuGraphicsResourceGetMappedPointer_v2;
+tcuGraphicsResourceSetMapFlags_v2 *cuGraphicsResourceSetMapFlags_v2;
+tcuGraphicsMapResources *cuGraphicsMapResources;
+tcuGraphicsUnmapResources *cuGraphicsUnmapResources;
+tcuGetExportTable *cuGetExportTable;
+
+tcuGraphicsGLRegisterBuffer *cuGraphicsGLRegisterBuffer;
+tcuGraphicsGLRegisterImage *cuGraphicsGLRegisterImage;
+tcuGLGetDevices_v2 *cuGLGetDevices_v2;
+tcuGLCtxCreate_v2 *cuGLCtxCreate_v2;
+tcuGLInit *cuGLInit;
+tcuGLRegisterBufferObject *cuGLRegisterBufferObject;
+tcuGLMapBufferObject_v2 *cuGLMapBufferObject_v2;
+tcuGLUnmapBufferObject *cuGLUnmapBufferObject;
+tcuGLUnregisterBufferObject *cuGLUnregisterBufferObject;
+tcuGLSetBufferObjectMapFlags *cuGLSetBufferObjectMapFlags;
+tcuGLMapBufferObjectAsync_v2 *cuGLMapBufferObjectAsync_v2;
+tcuGLUnmapBufferObjectAsync *cuGLUnmapBufferObjectAsync;
+
+tcuImportExternalMemory *cuImportExternalMemory;
+tcuExternalMemoryGetMappedBuffer *cuExternalMemoryGetMappedBuffer;
+tcuExternalMemoryGetMappedMipmappedArray *cuExternalMemoryGetMappedMipmappedArray;
+tcuDestroyExternalMemory *cuDestroyExternalMemory;
+
+tnvrtcGetErrorString *nvrtcGetErrorString;
+tnvrtcVersion *nvrtcVersion;
+tnvrtcCreateProgram *nvrtcCreateProgram;
+tnvrtcDestroyProgram *nvrtcDestroyProgram;
+tnvrtcCompileProgram *nvrtcCompileProgram;
+tnvrtcGetPTXSize *nvrtcGetPTXSize;
+tnvrtcGetPTX *nvrtcGetPTX;
+tnvrtcGetProgramLogSize *nvrtcGetProgramLogSize;
+tnvrtcGetProgramLog *nvrtcGetProgramLog;
+tnvrtcAddNameExpression *nvrtcAddNameExpression;
+tnvrtcGetLoweredName *nvrtcGetLoweredName;
+
+
+static DynamicLibrary dynamic_library_open_find(const char **paths) {
+  int i = 0;
+  while (paths[i] != NULL) {
+      DynamicLibrary lib = dynamic_library_open(paths[i]);
+      if (lib != NULL) {
+        return lib;
+      }
+      ++i;
+  }
+  return NULL;
+}
+
+/* Implementation function. */
+static void cuewCudaExit(void) {
+  if (cuda_lib != NULL) {
+    /*  Ignore errors. */
+    dynamic_library_close(cuda_lib);
+    cuda_lib = NULL;
+  }
+}
+
+static int cuewCudaInit(void) {
+  /* Library paths. */
+#ifdef _WIN32
+  /* Expected in c:/windows/system or similar, no path needed. */
+  const char *cuda_paths[] = {"nvcuda.dll", NULL};
+#elif defined(__APPLE__)
+  /* Default installation path. */
+  const char *cuda_paths[] = {"/usr/local/cuda/lib/libcuda.dylib", NULL};
+#else
+  const char *cuda_paths[] = {"libcuda.so", "libcuda.so.1", NULL};
+#endif
+  static int initialized = 0;
+  static int result = 0;
+  int error, driver_version;
+
+  if (initialized) {
+    return result;
+  }
+
+  initialized = 1;
+
+  error = atexit(cuewCudaExit);
+  if (error) {
+    result = CUEW_ERROR_ATEXIT_FAILED;
+    return result;
+  }
+
+  /* Load library. */
+  cuda_lib = dynamic_library_open_find(cuda_paths);
+
+  if (cuda_lib == NULL) {
+    result = CUEW_ERROR_OPEN_FAILED;
+    return result;
+  }
+
+  /* Detect driver version. */
+  driver_version = 1000;
+
+  CUDA_LIBRARY_FIND_CHECKED(cuDriverGetVersion);
+  if (cuDriverGetVersion) {
+    cuDriverGetVersion(&driver_version);
+  }
+
+  /* We require version 4.0. */
+  if (driver_version < 4000) {
+    result = CUEW_ERROR_OPEN_FAILED;
+    return result;
+  }
+  /* Fetch all function pointers. */
+  CUDA_LIBRARY_FIND(cuGetErrorString);
+  CUDA_LIBRARY_FIND(cuGetErrorName);
+  CUDA_LIBRARY_FIND(cuInit);
+  CUDA_LIBRARY_FIND(cuDriverGetVersion);
+  CUDA_LIBRARY_FIND(cuDeviceGet);
+  CUDA_LIBRARY_FIND(cuDeviceGetCount);
+  CUDA_LIBRARY_FIND(cuDeviceGetName);
+  CUDA_LIBRARY_FIND(cuDeviceGetUuid);
+  CUDA_LIBRARY_FIND(cuDeviceTotalMem_v2);
+  CUDA_LIBRARY_FIND(cuDeviceGetAttribute);
+  CUDA_LIBRARY_FIND(cuDeviceGetProperties);
+  CUDA_LIBRARY_FIND(cuDeviceComputeCapability);
+  CUDA_LIBRARY_FIND(cuDevicePrimaryCtxRetain);
+  CUDA_LIBRARY_FIND(cuDevicePrimaryCtxRelease);
+  CUDA_LIBRARY_FIND(cuDevicePrimaryCtxSetFlags);
+  CUDA_LIBRARY_FIND(cuDevicePrimaryCtxGetState);
+  CUDA_LIBRARY_FIND(cuDevicePrimaryCtxReset);
+  CUDA_LIBRARY_FIND(cuCtxCreate_v2);
+  CUDA_LIBRARY_FIND(cuCtxDestroy_v2);
+  CUDA_LIBRARY_FIND(cuCtxPushCurrent_v2);
+  CUDA_LIBRARY_FIND(cuCtxPopCurrent_v2);
+  CUDA_LIBRARY_FIND(cuCtxSetCurrent);
+  CUDA_LIBRARY_FIND(cuCtxGetCurrent);
+  CUDA_LIBRARY_FIND(cuCtxGetDevice);
+  CUDA_LIBRARY_FIND(cuCtxGetFlags);
+  CUDA_LIBRARY_FIND(cuCtxSynchronize);
+  CUDA_LIBRARY_FIND(cuCtxSetLimit);
+  CUDA_LIBRARY_FIND(cuCtxGetLimit);
+  CUDA_LIBRARY_FIND(cuCtxGetCacheConfig);
+  CUDA_LIBRARY_FIND(cuCtxSetCacheConfig);
+  CUDA_LIBRARY_FIND(cuCtxGetSharedMemConfig);
+  CUDA_LIBRARY_FIND(cuCtxSetSharedMemConfig);
+  CUDA_LIBRARY_FIND(cuCtxGetApiVersion);
+  CUDA_LIBRARY_FIND(cuCtxGetStreamPriorityRange);
+  CUDA_LIBRARY_FIND(cuCtxAttach);
+  CUDA_LIBRARY_FIND(cuCtxDetach);
+  CUDA_LIBRARY_FIND(cuModuleLoad);
+  CUDA_LIBRARY_FIND(cuModuleLoadData);
+  CUDA_LIBRARY_FIND(cuModuleLoadDataEx);
+  CUDA_LIBRARY_FIND(cuModuleLoadFatBinary);
+  CUDA_LIBRARY_FIND(cuModuleUnload);
+  CUDA_LIBRARY_FIND(cuModuleGetFunction);
+  CUDA_LIBRARY_FIND(cuModuleGetGlobal_v2);
+  CUDA_LIBRARY_FIND(cuModuleGetTexRef);
+  CUDA_LIBRARY_FIND(cuModuleGetSurfRef);
+  CUDA_LIBRARY_FIND(cuLinkCreate_v2);
+  CUDA_LIBRARY_FIND(cuLinkAddData_v2);
+  CUDA_LIBRARY_FIND(cuLinkAddFile_v2);
+  CUDA_LIBRARY_FIND(cuLinkComplete);
+  CUDA_LIBRARY_FIND(cuLinkDestroy);
+  CUDA_LIBRARY_FIND(cuMemGetInfo_v2);
+  CUDA_LIBRARY_FIND(cuMemAlloc_v2);
+  CUDA_LIBRARY_FIND(cuMemAllocPitch_v2);
+  CUDA_LIBRARY_FIND(cuMemFree_v2);
+  CUDA_LIBRARY_FIND(cuMemGetAddressRange_v2);
+  CUDA_LIBRARY_FIND(cuMemAllocHost_v2);
+  CUDA_LIBRARY_FIND(cuMemFreeHost);
+  CUDA_LIBRARY_FIND(cuMemHostAlloc);
+  CUDA_LIBRARY_FIND(cuMemHostGetDevicePointer_v2);
+  CUDA_LIBRARY_FIND(cuMemHostGetFlags);
+  CUDA_LIBRARY_FIND(cuMemAllocManaged);
+  CUDA_LIBRARY_FIND(cuDeviceGetByPCIBusId);
+  CUDA_LIBRARY_FIND(cuDeviceGetPCIBusId);
+  CUDA_LIBRARY_FIND(cuIpcGetEventHandle);
+  CUDA_LIBRARY_FIND(cuIpcOpenEventHandle);
+  CUDA_LIBRARY_FIND(cuIpcGetMemHandle);
+  CUDA_LIBRARY_FIND(cuIpcOpenMemHandle);
+  CUDA_LIBRARY_FIND(cuIpcCloseMemHandle);
+  CUDA_LIBRARY_FIND(cuMemHostRegister_v2);
+  CUDA_LIBRARY_FIND(cuMemHostUnregister);
+  CUDA_LIBRARY_FIND(cuMemcpy);
+  CUDA_LIBRARY_FIND(cuMemcpyPeer);
+  CUDA_LIBRARY_FIND(cuMemcpyHtoD_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyDtoH_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyDtoD_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyDtoA_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyAtoD_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyHtoA_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyAtoH_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyAtoA_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy2D_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy2DUnaligned_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy3D_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy3DPeer);
+  CUDA_LIBRARY_FIND(cuMemcpyAsync);
+  CUDA_LIBRARY_FIND(cuMemcpyPeerAsync);
+  CUDA_LIBRARY_FIND(cuMemcpyHtoDAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyDtoHAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyDtoDAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyHtoAAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpyAtoHAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy2DAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy3DAsync_v2);
+  CUDA_LIBRARY_FIND(cuMemcpy3DPeerAsync);
+  CUDA_LIBRARY_FIND(cuMemsetD8_v2);
+  CUDA_LIBRARY_FIND(cuMemsetD16_v2);
+  CUDA_LIBRARY_FIND(cuMemsetD32_v2);
+  CUDA_LIBRARY_FIND(cuMemsetD2D8_v2);
+  CUDA_LIBRARY_FIND(cuMemsetD2D16_v2);
+  CUDA_LIBRARY_FIND(cuMemsetD2D32_v2);
+  CUDA_LIBRARY_FIND(cuMemsetD8Async);
+  CUDA_LIBRARY_FIND(cuMemsetD16Async);
+  CUDA_LIBRARY_FIND(cuMemsetD32Async);
+  CUDA_LIBRARY_FIND(cuMemsetD2D8Async);
+  CUDA_LIBRARY_FIND(cuMemsetD2D16Async);
+  CUDA_LIBRARY_FIND(cuMemsetD2D32Async);
+  CUDA_LIBRARY_FIND(cuArrayCreate_v2);
+  CUDA_LIBRARY_FIND(cuArrayGetDescriptor_v2);
+  CUDA_LIBRARY_FIND(cuArrayDestroy);
+  CUDA_LIBRARY_FIND(cuArray3DCreate_v2);
+  CUDA_LIBRARY_FIND(cuArray3DGetDescriptor_v2);
+  CUDA_LIBRARY_FIND(cuMipmappedArrayCreate);
+  CUDA_LIBRARY_FIND(cuMipmappedArrayGetLevel);
+  CUDA_LIBRARY_FIND(cuMipmappedArrayDestroy);
+  CUDA_LIBRARY_FIND(cuPointerGetAttribute);
+  CUDA_LIBRARY_FIND(cuMemPrefetchAsync);
+  CUDA_LIBRARY_FIND(cuMemAdvise);
+  CUDA_LIBRARY_FIND(cuMemRangeGetAttribute);
+  CUDA_LIBRARY_FIND(cuMemRangeGetAttributes);
+  CUDA_LIBRARY_FIND(cuPointerSetAttribute);
+  CUDA_LIBRARY_FIND(cuPointerGetAttributes);
+  CUDA_LIBRARY_FIND(cuStreamCreate);
+  CUDA_LIBRARY_FIND(cuStreamCreateWithPriority);
+  CUDA_LIBRARY_FIND(cuStreamGetPriority);
+  CUDA_LIBRARY_FIND(cuStreamGetFlags);
+  CUDA_LIBRARY_FIND(cuStreamGetCtx);
+  CUDA_LIBRARY_FIND(cuStreamWaitEvent);
+  CUDA_LIBRARY_FIND(cuStreamAddCallback);
+  CUDA_LIBRARY_FIND(cuStreamAttachMemAsync);
+  CUDA_LIBRARY_FIND(cuStreamQuery);
+  CUDA_LIBRARY_FIND(cuStreamSynchronize);
+  CUDA_LIBRARY_FIND(cuStreamDestroy_v2);
+  CUDA_LIBRARY_FIND(cuEventCreate);
+  CUDA_LIBRARY_FIND(cuEventRecord);
+  CUDA_LIBRARY_FIND(cuEventQuery);
+  CUDA_LIBRARY_FIND(cuEventSynchronize);
+  CUDA_LIBRARY_FIND(cuEventDestroy_v2);
+  CUDA_LIBRARY_FIND(cuEventElapsedTime);
+  CUDA_LIBRARY_FIND(cuStreamWaitValue32);
+  CUDA_LIBRARY_FIND(cuStreamWaitValue64);
+  CUDA_LIBRARY_FIND(cuStreamWriteValue32);
+  CUDA_LIBRARY_FIND(cuStreamWriteValue64);
+  CUDA_LIBRARY_FIND(cuStreamBatchMemOp);
+  CUDA_LIBRARY_FIND(cuFuncGetAttribute);
+  CUDA_LIBRARY_FIND(cuFuncSetAttribute);
+  CUDA_LIBRARY_FIND(cuFuncSetCacheConfig);
+  CUDA_LIBRARY_FIND(cuFuncSetSharedMemConfig);
+  CUDA_LIBRARY_FIND(cuLaunchKernel);
+  CUDA_LIBRARY_FIND(cuLaunchCooperativeKernel);
+  CUDA_LIBRARY_FIND(cuLaunchCooperativeKernelMultiDevice);
+  CUDA_LIBRARY_FIND(cuFuncSetBlockShape);
+  CUDA_LIBRARY_FIND(cuFuncSetSharedSize);
+  CUDA_LIBRARY_FIND(cuParamSetSize);
+  CUDA_LIBRARY_FIND(cuParamSeti);
+  CUDA_LIBRARY_FIND(cuParamSetf);
+  CUDA_LIBRARY_FIND(cuParamSetv);
+  CUDA_LIBRARY_FIND(cuLaunch);
+  CUDA_LIBRARY_FIND(cuLaunchGrid);
+  CUDA_LIBRARY_FIND(cuLaunchGridAsync);
+  CUDA_LIBRARY_FIND(cuParamSetTexRef);
+  CUDA_LIBRARY_FIND(cuOccupancyMaxActiveBlocksPerMultiprocessor);
+  CUDA_LIBRARY_FIND(cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags);
+  CUDA_LIBRARY_FIND(cuOccupancyMaxPotentialBlockSize);
+  CUDA_LIBRARY_FIND(cuOccupancyMaxPotentialBlockSizeWithFlags);
+  CUDA_LIBRARY_FIND(cuTexRefSetArray);
+  CUDA_LIBRARY_FIND(cuTexRefSetMipmappedArray);
+  CUDA_LIBRARY_FIND(cuTexRefSetAddress_v2);
+  CUDA_LIBRARY_FIND(cuTexRefSetAddress2D_v3);
+  CUDA_LIBRARY_FIND(cuTexRefSetFormat);
+  CUDA_LIBRARY_FIND(cuTexRefSetAddressMode);
+  CUDA_LIBRARY_FIND(cuTexRefSetFilterMode);
+  CUDA_LIBRARY_FIND(cuTexRefSetMipmapFilterMode);
+  CUDA_LIBRARY_FIND(cuTexRefSetMipmapLevelBias);
+  CUDA_LIBRARY_FIND(cuTexRefSetMipmapLevelClamp);
+  CUDA_LIBRARY_FIND(cuTexRefSetMaxAnisotropy);
+  CUDA_LIBRARY_FIND(cuTexRefSetBorderColor);
+  CUDA_LIBRARY_FIND(cuTexRefSetFlags);
+  CUDA_LIBRARY_FIND(cuTexRefGetAddress_v2);
+  CUDA_LIBRARY_FIND(cuTexRefGetArray);
+  CUDA_LIBRARY_FIND(cuTexRefGetMipmappedArray);
+  CUDA_LIBRARY_FIND(cuTexRefGetAddressMode);
+  CUDA_LIBRARY_FIND(cuTexRefGetFilterMode);
+  CUDA_LIBRARY_FIND(cuTexRefGetFormat);
+  CUDA_LIBRARY_FIND(cuTexRefGetMipmapFilterMode);
+  CUDA_LIBRARY_FIND(cuTexRefGetMipmapLevelBias);
+  CUDA_LIBRARY_FIND(cuTexRefGetMipmapLevelClamp);
+  CUDA_LIBRARY_FIND(cuTexRefGetMaxAnisotropy);
+  CUDA_LIBRARY_FIND(cuTexRefGetBorderColor);
+  CUDA_LIBRARY_FIND(cuTexRefGetFlags);
+  CUDA_LIBRARY_FIND(cuTexRefCreate);
+  CUDA_LIBRARY_FIND(cuTexRefDestroy);
+  CUDA_LIBRARY_FIND(cuSurfRefSetArray);
+  CUDA_LIBRARY_FIND(cuSurfRefGetArray);
+  CUDA_LIBRARY_FIND(cuTexObjectCreate);
+  CUDA_LIBRARY_FIND(cuTexObjectDestroy);
+  CUDA_LIBRARY_FIND(cuTexObjectGetResourceDesc);
+  CUDA_LIBRARY_FIND(cuTexObjectGetTextureDesc);
+  CUDA_LIBRARY_FIND(cuTexObjectGetResourceViewDesc);
+  CUDA_LIBRARY_FIND(cuSurfObjectCreate);
+  CUDA_LIBRARY_FIND(cuSurfObjectDestroy);
+  CUDA_LIBRARY_FIND(cuSurfObjectGetResourceDesc);
+  CUDA_LIBRARY_FIND(cuDeviceCanAccessPeer);
+  CUDA_LIBRARY_FIND(cuCtxEnablePeerAccess);
+  CUDA_LIBRARY_FIND(cuCtxDisablePeerAccess);
+  CUDA_LIBRARY_FIND(cuDeviceGetP2PAttribute);
+  CUDA_LIBRARY_FIND(cuGraphicsUnregisterResource);
+  CUDA_LIBRARY_FIND(cuGraphicsSubResourceGetMappedArray);
+  CUDA_LIBRARY_FIND(cuGraphicsResourceGetMappedMipmappedArray);
+  CUDA_LIBRARY_FIND(cuGraphicsResourceGetMappedPointer_v2);
+  CUDA_LIBRARY_FIND(cuGraphicsResourceSetMapFlags_v2);
+  CUDA_LIBRARY_FIND(cuGraphicsMapResources);
+  CUDA_LIBRARY_FIND(cuGraphicsUnmapResources);
+  CUDA_LIBRARY_FIND(cuGetExportTable);
+
+  CUDA_LIBRARY_FIND(cuGraphicsGLRegisterBuffer);
+  CUDA_LIBRARY_FIND(cuGraphicsGLRegisterImage);
+  CUDA_LIBRARY_FIND(cuGLGetDevices_v2);
+  CUDA_LIBRARY_FIND(cuGLCtxCreate_v2);
+  CUDA_LIBRARY_FIND(cuGLInit);
+  CUDA_LIBRARY_FIND(cuGLRegisterBufferObject);
+  CUDA_LIBRARY_FIND(cuGLMapBufferObject_v2);
+  CUDA_LIBRARY_FIND(cuGLUnmapBufferObject);
+  CUDA_LIBRARY_FIND(cuGLUnregisterBufferObject);
+  CUDA_LIBRARY_FIND(cuGLSetBufferObjectMapFlags);
+  CUDA_LIBRARY_FIND(cuGLMapBufferObjectAsync_v2);
+  CUDA_LIBRARY_FIND(cuGLUnmapBufferObjectAsync);
+
+  CUDA_LIBRARY_FIND(cuImportExternalMemory);
+  CUDA_LIBRARY_FIND(cuExternalMemoryGetMappedBuffer);
+  CUDA_LIBRARY_FIND(cuExternalMemoryGetMappedMipmappedArray);
+  CUDA_LIBRARY_FIND(cuDestroyExternalMemory);
+
+  result = CUEW_SUCCESS;
+  return result;
+}
+
+static void cuewExitNvrtc(void) {
+  if (nvrtc_lib != NULL) {
+    /*  Ignore errors. */
+    dynamic_library_close(nvrtc_lib);
+    nvrtc_lib = NULL;
+  }
+}
+
+static int cuewNvrtcInit(void) {
+  /* Library paths. */
+#ifdef _WIN32
+  /* Expected in c:/windows/system or similar, no path needed. */
+  const char *nvrtc_paths[] = {"nvrtc64_101_0.dll",
+                               "nvrtc64_100_0.dll",
+                               "nvrtc64_91.dll",
+                               "nvrtc64_90.dll",
+                               "nvrtc64_80.dll",
+                               NULL};
+#elif defined(__APPLE__)
+  /* Default installation path. */
+  const char *nvrtc_paths[] = {"/usr/local/cuda/lib/libnvrtc.dylib", NULL};
+#else
+  const char *nvrtc_paths[] = {"libnvrtc.so",
+#  if defined(__x86_64__) || defined(_M_X64)
+                               "/usr/local/cuda/lib64/libnvrtc.so",
+#else
+                               "/usr/local/cuda/lib/libnvrtc.so",
+#endif
+                               NULL};
+#endif
+  static int initialized = 0;
+  static int result = 0;
+  int error;
+
+  if (initialized) {
+    return result;
+  }
+
+  initialized = 1;
+
+  error = atexit(cuewExitNvrtc);
+  if (error) {
+    result = CUEW_ERROR_ATEXIT_FAILED;
+    return result;
+  }
+
+  /* Load library. */
+  nvrtc_lib = dynamic_library_open_find(nvrtc_paths);
+
+  if (nvrtc_lib == NULL) {
+    result = CUEW_ERROR_OPEN_FAILED;
+    return result;
+  }
+
+  NVRTC_LIBRARY_FIND(nvrtcGetErrorString);
+  NVRTC_LIBRARY_FIND(nvrtcVersion);
+  NVRTC_LIBRARY_FIND(nvrtcCreateProgram);
+  NVRTC_LIBRARY_FIND(nvrtcDestroyProgram);
+  NVRTC_LIBRARY_FIND(nvrtcCompileProgram);
+  NVRTC_LIBRARY_FIND(nvrtcGetPTXSize);
+  NVRTC_LIBRARY_FIND(nvrtcGetPTX);
+  NVRTC_LIBRARY_FIND(nvrtcGetProgramLogSize);
+  NVRTC_LIBRARY_FIND(nvrtcGetProgramLog);
+  NVRTC_LIBRARY_FIND(nvrtcAddNameExpression);
+  NVRTC_LIBRARY_FIND(nvrtcGetLoweredName);
+
+  result = CUEW_SUCCESS;
+  return result;
+}
+
+
+int cuewInit(cuuint32_t flags) {
+  int result = CUEW_SUCCESS;
+
+  if (flags & CUEW_INIT_CUDA) {
+    result = cuewCudaInit();
+    if (result != CUEW_SUCCESS) {
+      return result;
+    }
+  }
+
+  if (flags & CUEW_INIT_NVRTC) {
+    result = cuewNvrtcInit();
+    if (result != CUEW_SUCCESS) {
+      return result;
+    }
+  }
+
+  return result;
+}
+
+
+const char *cuewErrorString(CUresult result) {
+  switch (result) {
+    case CUDA_SUCCESS: return "No errors";
+    case CUDA_ERROR_INVALID_VALUE: return "Invalid value";
+    case CUDA_ERROR_OUT_OF_MEMORY: return "Out of memory";
+    case CUDA_ERROR_NOT_INITIALIZED: return "Driver not initialized";
+    case CUDA_ERROR_DEINITIALIZED: return "Driver deinitialized";
+    case CUDA_ERROR_PROFILER_DISABLED: return "Profiler disabled";
+    case CUDA_ERROR_PROFILER_NOT_INITIALIZED: return "Profiler not initialized";
+    case CUDA_ERROR_PROFILER_ALREADY_STARTED: return "Profiler already started";
+    case CUDA_ERROR_PROFILER_ALREADY_STOPPED: return "Profiler already stopped";
+    case CUDA_ERROR_NO_DEVICE: return "No CUDA-capable device available";
+    case CUDA_ERROR_INVALID_DEVICE: return "Invalid device";
+    case CUDA_ERROR_INVALID_IMAGE: return "Invalid kernel image";
+    case CUDA_ERROR_INVALID_CONTEXT: return "Invalid context";
+    case CUDA_ERROR_CONTEXT_ALREADY_CURRENT: return "Context already current";
+    case CUDA_ERROR_MAP_FAILED: return "Map failed";
+    case CUDA_ERROR_UNMAP_FAILED: return "Unmap failed";
+    case CUDA_ERROR_ARRAY_IS_MAPPED: return "Array is mapped";
+    case CUDA_ERROR_ALREADY_MAPPED: return "Already mapped";
+    case CUDA_ERROR_NO_BINARY_FOR_GPU: return "No binary for GPU";
+    case CUDA_ERROR_ALREADY_ACQUIRED: return "Already acquired";
+    case CUDA_ERROR_NOT_MAPPED: return "Not mapped";
+    case CUDA_ERROR_NOT_MAPPED_AS_ARRAY: return "Mapped resource not available for access as an array";
+    case CUDA_ERROR_NOT_MAPPED_AS_POINTER: return "Mapped resource not available for access as a pointer";
+    case CUDA_ERROR_ECC_UNCORRECTABLE: return "Uncorrectable ECC error detected";
+    case CUDA_ERROR_UNSUPPORTED_LIMIT: return "CUlimit not supported by device";
+    case CUDA_ERROR_CONTEXT_ALREADY_IN_USE: return "Context already in use";
+    case CUDA_ERROR_PEER_ACCESS_UNSUPPORTED: return "Peer access unsupported";
+    case CUDA_ERROR_INVALID_PTX: return "Invalid ptx";
+    case CUDA_ERROR_INVALID_GRAPHICS_CONTEXT: return "Invalid graphics context";
+    case CUDA_ERROR_NVLINK_UNCORRECTABLE: return "Nvlink uncorrectable";
+    case CUDA_ERROR_JIT_COMPILER_NOT_FOUND: return "Jit compiler not found";
+    case CUDA_ERROR_UNSUPPORTED_PTX_VERSION: return "Unsupported PTX version";
+    case CUDA_ERROR_INVALID_SOURCE: return "Invalid source";
+    case CUDA_ERROR_FILE_NOT_FOUND: return "File not found";
+    case CUDA_ERROR_SHARED_OBJECT_SYMBOL_NOT_FOUND: return "Link to a shared object failed to resolve";
+    case CUDA_ERROR_SHARED_OBJECT_INIT_FAILED: return "Shared object initialization failed";
+    case CUDA_ERROR_OPERATING_SYSTEM: return "Operating system";
+    case CUDA_ERROR_INVALID_HANDLE: return "Invalid handle";
+    case CUDA_ERROR_NOT_FOUND: return "Not found";
+    case CUDA_ERROR_NOT_READY: return "CUDA not ready";
+    case CUDA_ERROR_ILLEGAL_ADDRESS: return "Illegal address";
+    case CUDA_ERROR_LAUNCH_OUT_OF_RESOURCES: return "Launch exceeded resources";
+    case CUDA_ERROR_LAUNCH_TIMEOUT: return "Launch exceeded timeout";
+    case CUDA_ERROR_LAUNCH_INCOMPATIBLE_TEXTURING: return "Launch with incompatible texturing";
+    case CUDA_ERROR_PEER_ACCESS_ALREADY_ENABLED: return "Peer access already enabled";
+    case CUDA_ERROR_PEER_ACCESS_NOT_ENABLED: return "Peer access not enabled";
+    case CUDA_ERROR_PRIMARY_CONTEXT_ACTIVE: return "Primary context active";
+    case CUDA_ERROR_CONTEXT_IS_DESTROYED: return "Context is destroyed";
+    case CUDA_ERROR_ASSERT: return "Assert";
+    case CUDA_ERROR_TOO_MANY_PEERS: return "Too many peers";
+    case CUDA_ERROR_HOST_MEMORY_ALREADY_REGISTERED: return "Host memory already registered";
+    case CUDA_ERROR_HOST_MEMORY_NOT_REGISTERED: return "Host memory not registered";
+    case CUDA_ERROR_HARDWARE_STACK_ERROR: return "Hardware stack error";
+    case CUDA_ERROR_ILLEGAL_INSTRUCTION: return "Illegal instruction";
+    case CUDA_ERROR_MISALIGNED_ADDRESS: return "Misaligned address";
+    case CUDA_ERROR_INVALID_ADDRESS_SPACE: return "Invalid address space";
+    case CUDA_ERROR_INVALID_PC: return "Invalid pc";
+    case CUDA_ERROR_LAUNCH_FAILED: return "Launch failed";
+    case CUDA_ERROR_COOPERATIVE_LAUNCH_TOO_LARGE: return "Cooperative launch too large";
+    case CUDA_ERROR_NOT_PERMITTED: return "Not permitted";
+    case CUDA_ERROR_NOT_SUPPORTED: return "Not supported";
+    case CUDA_ERROR_UNKNOWN: return "Unknown error";
+    default: return "Unknown CUDA error value";
+  }
+}
+
+static void path_join(const char *path1,
+                      const char *path2,
+                      int maxlen,
+                      char *result) {
+#if defined(WIN32) || defined(_WIN32)
+  const char separator = '\\';
+#else
+  const char separator = '/';
+#endif
+  int n = snprintf(result, maxlen, "%s%c%s", path1, separator, path2);
+  if (n != -1 && n < maxlen) {
+    result[n] = '\0';
+  }
+  else {
+    result[maxlen - 1] = '\0';
+  }
+}
+
+static int path_exists(const char *path) {
+  struct stat st;
+  if (stat(path, &st)) {
+    return 0;
+  }
+  return 1;
+}
+
+const char *cuewCompilerPath(void) {
+#ifdef _WIN32
+  const char *defaultpaths[] = {
+    "C:/CUDA/bin",
+    "C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.1/bin",
+    NULL};
+  const char *executable = "nvcc.exe";
+#else
+  const char *defaultpaths[] = {
+    "/Developer/NVIDIA/CUDA-5.0/bin",
+    "/usr/local/cuda-5.0/bin",
+    "/usr/local/cuda/bin",
+    "/Developer/NVIDIA/CUDA-6.0/bin",
+    "/usr/local/cuda-6.0/bin",
+    "/Developer/NVIDIA/CUDA-5.5/bin",
+    "/usr/local/cuda-5.5/bin",
+    NULL};
+  const char *executable = "nvcc";
+#endif
+  int i;
+
+  const char *binpath = getenv("CUDA_BIN_PATH");
+
+  static char nvcc[65536];
+
+  if (binpath) {
+    path_join(binpath, executable, sizeof(nvcc), nvcc);
+    if (path_exists(nvcc)) {
+      return nvcc;
+    }
+  }
+
+  for (i = 0; defaultpaths[i]; ++i) {
+    path_join(defaultpaths[i], executable, sizeof(nvcc), nvcc);
+    if (path_exists(nvcc)) {
+      return nvcc;
+    }
+  }
+
+  {
+#ifdef _WIN32
+    FILE *handle = popen("where nvcc", "r");
+#else
+    FILE *handle = popen("which nvcc", "r");
+#endif
+    if (handle) {
+      char buffer[4096] = {0};
+      int len = fread(buffer, 1, sizeof(buffer) - 1, handle);
+      buffer[len] = '\0';
+      pclose(handle);
+      if (buffer[0]) {
+        return "nvcc";
+      }
+    }
+  }
+
+  return NULL;
+}
+
+int cuewNvrtcVersion(void) {
+  int major, minor;
+  if (nvrtcVersion) {
+    nvrtcVersion(&major, &minor);
+    return 10 * major + minor;
+  }
+  return 0;
+}
+
+int cuewCompilerVersion(void) {
+  const char *path = cuewCompilerPath();
+  const char *marker = "Cuda compilation tools, release ";
+  FILE *pipe;
+  int major, minor;
+  char *versionstr;
+  char buf[128];
+  char output[65536] = "\0";
+  char command[65536] = "\0";
+
+  if (path == NULL) {
+    return 0;
+  }
+
+  /* get --version output */
+  strcat(command, "\"");
+  strncat(command, path, sizeof(command) - 1);
+  strncat(command, "\" --version", sizeof(command) - strlen(path) - 1);
+  pipe = popen(command, "r");
+  if (!pipe) {
+    fprintf(stderr, "CUDA: failed to run compiler to retrieve version");
+    return 0;
+  }
+
+  while (!feof(pipe)) {
+    if (fgets(buf, sizeof(buf), pipe) != NULL) {
+      strncat(output, buf, sizeof(output) - strlen(output) - 1);
+    }
+  }
+
+  pclose(pipe);
+
+  /* parse version number */
+  versionstr = strstr(output, marker);
+  if (versionstr == NULL) {
+    fprintf(stderr, "CUDA: failed to find version number in:\n\n%s\n", output);
+    return 0;
+  }
+  versionstr += strlen(marker);
+
+  if (sscanf(versionstr, "%d.%d", &major, &minor) < 2) {
+    fprintf(stderr, "CUDA: failed to parse version number from:\n\n%s\n", output);
+    return 0;
+  }
+
+  return 10 * major + minor;
+}