Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#ifndef VULKAN_WIN32_H_
|
|
|
|
#define VULKAN_WIN32_H_ 1
|
|
|
|
|
|
|
|
/*
|
|
|
|
** Copyright 2015-2023 The Khronos Group Inc.
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
**
|
|
|
|
** SPDX-License-Identifier: Apache-2.0
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
** This header is generated from the Khronos Vulkan XML API Registry.
|
|
|
|
**
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
|
|
#ifdef __cplusplus
|
|
|
|
extern "C" {
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_KHR_win32_surface 1
|
|
|
|
#define VK_KHR_WIN32_SURFACE_SPEC_VERSION 6
|
|
|
|
#define VK_KHR_WIN32_SURFACE_EXTENSION_NAME "VK_KHR_win32_surface"
|
|
|
|
typedef VkFlags VkWin32SurfaceCreateFlagsKHR;
|
|
|
|
typedef struct VkWin32SurfaceCreateInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkWin32SurfaceCreateFlagsKHR flags;
|
|
|
|
HINSTANCE hinstance;
|
|
|
|
HWND hwnd;
|
|
|
|
} VkWin32SurfaceCreateInfoKHR;
|
|
|
|
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkCreateWin32SurfaceKHR)(VkInstance instance, const VkWin32SurfaceCreateInfoKHR* pCreateInfo, const VkAllocationCallbacks* pAllocator, VkSurfaceKHR* pSurface);
|
|
|
|
typedef VkBool32 (VKAPI_PTR *PFN_vkGetPhysicalDeviceWin32PresentationSupportKHR)(VkPhysicalDevice physicalDevice, uint32_t queueFamilyIndex);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkCreateWin32SurfaceKHR(
|
|
|
|
VkInstance instance,
|
|
|
|
const VkWin32SurfaceCreateInfoKHR* pCreateInfo,
|
|
|
|
const VkAllocationCallbacks* pAllocator,
|
|
|
|
VkSurfaceKHR* pSurface);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkBool32 VKAPI_CALL vkGetPhysicalDeviceWin32PresentationSupportKHR(
|
|
|
|
VkPhysicalDevice physicalDevice,
|
|
|
|
uint32_t queueFamilyIndex);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_KHR_external_memory_win32 1
|
|
|
|
#define VK_KHR_EXTERNAL_MEMORY_WIN32_SPEC_VERSION 1
|
|
|
|
#define VK_KHR_EXTERNAL_MEMORY_WIN32_EXTENSION_NAME "VK_KHR_external_memory_win32"
|
|
|
|
typedef struct VkImportMemoryWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkExternalMemoryHandleTypeFlagBits handleType;
|
|
|
|
HANDLE handle;
|
|
|
|
LPCWSTR name;
|
|
|
|
} VkImportMemoryWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkExportMemoryWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
const SECURITY_ATTRIBUTES* pAttributes;
|
|
|
|
DWORD dwAccess;
|
|
|
|
LPCWSTR name;
|
|
|
|
} VkExportMemoryWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkMemoryWin32HandlePropertiesKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
void* pNext;
|
|
|
|
uint32_t memoryTypeBits;
|
|
|
|
} VkMemoryWin32HandlePropertiesKHR;
|
|
|
|
|
|
|
|
typedef struct VkMemoryGetWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkDeviceMemory memory;
|
|
|
|
VkExternalMemoryHandleTypeFlagBits handleType;
|
|
|
|
} VkMemoryGetWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetMemoryWin32HandleKHR)(VkDevice device, const VkMemoryGetWin32HandleInfoKHR* pGetWin32HandleInfo, HANDLE* pHandle);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetMemoryWin32HandlePropertiesKHR)(VkDevice device, VkExternalMemoryHandleTypeFlagBits handleType, HANDLE handle, VkMemoryWin32HandlePropertiesKHR* pMemoryWin32HandleProperties);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetMemoryWin32HandleKHR(
|
|
|
|
VkDevice device,
|
|
|
|
const VkMemoryGetWin32HandleInfoKHR* pGetWin32HandleInfo,
|
|
|
|
HANDLE* pHandle);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetMemoryWin32HandlePropertiesKHR(
|
|
|
|
VkDevice device,
|
|
|
|
VkExternalMemoryHandleTypeFlagBits handleType,
|
|
|
|
HANDLE handle,
|
|
|
|
VkMemoryWin32HandlePropertiesKHR* pMemoryWin32HandleProperties);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_KHR_win32_keyed_mutex 1
|
|
|
|
#define VK_KHR_WIN32_KEYED_MUTEX_SPEC_VERSION 1
|
|
|
|
#define VK_KHR_WIN32_KEYED_MUTEX_EXTENSION_NAME "VK_KHR_win32_keyed_mutex"
|
|
|
|
typedef struct VkWin32KeyedMutexAcquireReleaseInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
uint32_t acquireCount;
|
|
|
|
const VkDeviceMemory* pAcquireSyncs;
|
|
|
|
const uint64_t* pAcquireKeys;
|
|
|
|
const uint32_t* pAcquireTimeouts;
|
|
|
|
uint32_t releaseCount;
|
|
|
|
const VkDeviceMemory* pReleaseSyncs;
|
|
|
|
const uint64_t* pReleaseKeys;
|
|
|
|
} VkWin32KeyedMutexAcquireReleaseInfoKHR;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#define VK_KHR_external_semaphore_win32 1
|
|
|
|
#define VK_KHR_EXTERNAL_SEMAPHORE_WIN32_SPEC_VERSION 1
|
|
|
|
#define VK_KHR_EXTERNAL_SEMAPHORE_WIN32_EXTENSION_NAME "VK_KHR_external_semaphore_win32"
|
|
|
|
typedef struct VkImportSemaphoreWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkSemaphore semaphore;
|
|
|
|
VkSemaphoreImportFlags flags;
|
|
|
|
VkExternalSemaphoreHandleTypeFlagBits handleType;
|
|
|
|
HANDLE handle;
|
|
|
|
LPCWSTR name;
|
|
|
|
} VkImportSemaphoreWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkExportSemaphoreWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
const SECURITY_ATTRIBUTES* pAttributes;
|
|
|
|
DWORD dwAccess;
|
|
|
|
LPCWSTR name;
|
|
|
|
} VkExportSemaphoreWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkD3D12FenceSubmitInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
uint32_t waitSemaphoreValuesCount;
|
|
|
|
const uint64_t* pWaitSemaphoreValues;
|
|
|
|
uint32_t signalSemaphoreValuesCount;
|
|
|
|
const uint64_t* pSignalSemaphoreValues;
|
|
|
|
} VkD3D12FenceSubmitInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkSemaphoreGetWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkSemaphore semaphore;
|
|
|
|
VkExternalSemaphoreHandleTypeFlagBits handleType;
|
|
|
|
} VkSemaphoreGetWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkImportSemaphoreWin32HandleKHR)(VkDevice device, const VkImportSemaphoreWin32HandleInfoKHR* pImportSemaphoreWin32HandleInfo);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetSemaphoreWin32HandleKHR)(VkDevice device, const VkSemaphoreGetWin32HandleInfoKHR* pGetWin32HandleInfo, HANDLE* pHandle);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkImportSemaphoreWin32HandleKHR(
|
|
|
|
VkDevice device,
|
|
|
|
const VkImportSemaphoreWin32HandleInfoKHR* pImportSemaphoreWin32HandleInfo);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetSemaphoreWin32HandleKHR(
|
|
|
|
VkDevice device,
|
|
|
|
const VkSemaphoreGetWin32HandleInfoKHR* pGetWin32HandleInfo,
|
|
|
|
HANDLE* pHandle);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_KHR_external_fence_win32 1
|
|
|
|
#define VK_KHR_EXTERNAL_FENCE_WIN32_SPEC_VERSION 1
|
|
|
|
#define VK_KHR_EXTERNAL_FENCE_WIN32_EXTENSION_NAME "VK_KHR_external_fence_win32"
|
|
|
|
typedef struct VkImportFenceWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkFence fence;
|
|
|
|
VkFenceImportFlags flags;
|
|
|
|
VkExternalFenceHandleTypeFlagBits handleType;
|
|
|
|
HANDLE handle;
|
|
|
|
LPCWSTR name;
|
|
|
|
} VkImportFenceWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkExportFenceWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
const SECURITY_ATTRIBUTES* pAttributes;
|
|
|
|
DWORD dwAccess;
|
|
|
|
LPCWSTR name;
|
|
|
|
} VkExportFenceWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef struct VkFenceGetWin32HandleInfoKHR {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkFence fence;
|
|
|
|
VkExternalFenceHandleTypeFlagBits handleType;
|
|
|
|
} VkFenceGetWin32HandleInfoKHR;
|
|
|
|
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkImportFenceWin32HandleKHR)(VkDevice device, const VkImportFenceWin32HandleInfoKHR* pImportFenceWin32HandleInfo);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetFenceWin32HandleKHR)(VkDevice device, const VkFenceGetWin32HandleInfoKHR* pGetWin32HandleInfo, HANDLE* pHandle);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkImportFenceWin32HandleKHR(
|
|
|
|
VkDevice device,
|
|
|
|
const VkImportFenceWin32HandleInfoKHR* pImportFenceWin32HandleInfo);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetFenceWin32HandleKHR(
|
|
|
|
VkDevice device,
|
|
|
|
const VkFenceGetWin32HandleInfoKHR* pGetWin32HandleInfo,
|
|
|
|
HANDLE* pHandle);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_NV_external_memory_win32 1
|
|
|
|
#define VK_NV_EXTERNAL_MEMORY_WIN32_SPEC_VERSION 1
|
|
|
|
#define VK_NV_EXTERNAL_MEMORY_WIN32_EXTENSION_NAME "VK_NV_external_memory_win32"
|
|
|
|
typedef struct VkImportMemoryWin32HandleInfoNV {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
VkExternalMemoryHandleTypeFlagsNV handleType;
|
|
|
|
HANDLE handle;
|
|
|
|
} VkImportMemoryWin32HandleInfoNV;
|
|
|
|
|
|
|
|
typedef struct VkExportMemoryWin32HandleInfoNV {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
const SECURITY_ATTRIBUTES* pAttributes;
|
|
|
|
DWORD dwAccess;
|
|
|
|
} VkExportMemoryWin32HandleInfoNV;
|
|
|
|
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetMemoryWin32HandleNV)(VkDevice device, VkDeviceMemory memory, VkExternalMemoryHandleTypeFlagsNV handleType, HANDLE* pHandle);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetMemoryWin32HandleNV(
|
|
|
|
VkDevice device,
|
|
|
|
VkDeviceMemory memory,
|
|
|
|
VkExternalMemoryHandleTypeFlagsNV handleType,
|
|
|
|
HANDLE* pHandle);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_NV_win32_keyed_mutex 1
|
|
|
|
#define VK_NV_WIN32_KEYED_MUTEX_SPEC_VERSION 2
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#define VK_NV_WIN32_KEYED_MUTEX_EXTENSION_NAME "VK_NV_win32_keyed_mutex"
|
|
|
|
typedef struct VkWin32KeyedMutexAcquireReleaseInfoNV {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
uint32_t acquireCount;
|
|
|
|
const VkDeviceMemory* pAcquireSyncs;
|
|
|
|
const uint64_t* pAcquireKeys;
|
|
|
|
const uint32_t* pAcquireTimeoutMilliseconds;
|
|
|
|
uint32_t releaseCount;
|
|
|
|
const VkDeviceMemory* pReleaseSyncs;
|
|
|
|
const uint64_t* pReleaseKeys;
|
|
|
|
} VkWin32KeyedMutexAcquireReleaseInfoNV;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#define VK_EXT_full_screen_exclusive 1
|
|
|
|
#define VK_EXT_FULL_SCREEN_EXCLUSIVE_SPEC_VERSION 4
|
|
|
|
#define VK_EXT_FULL_SCREEN_EXCLUSIVE_EXTENSION_NAME "VK_EXT_full_screen_exclusive"
|
|
|
|
|
|
|
|
typedef enum VkFullScreenExclusiveEXT {
|
|
|
|
VK_FULL_SCREEN_EXCLUSIVE_DEFAULT_EXT = 0,
|
|
|
|
VK_FULL_SCREEN_EXCLUSIVE_ALLOWED_EXT = 1,
|
|
|
|
VK_FULL_SCREEN_EXCLUSIVE_DISALLOWED_EXT = 2,
|
|
|
|
VK_FULL_SCREEN_EXCLUSIVE_APPLICATION_CONTROLLED_EXT = 3,
|
|
|
|
VK_FULL_SCREEN_EXCLUSIVE_MAX_ENUM_EXT = 0x7FFFFFFF
|
|
|
|
} VkFullScreenExclusiveEXT;
|
|
|
|
typedef struct VkSurfaceFullScreenExclusiveInfoEXT {
|
|
|
|
VkStructureType sType;
|
|
|
|
void* pNext;
|
|
|
|
VkFullScreenExclusiveEXT fullScreenExclusive;
|
|
|
|
} VkSurfaceFullScreenExclusiveInfoEXT;
|
|
|
|
|
|
|
|
typedef struct VkSurfaceCapabilitiesFullScreenExclusiveEXT {
|
|
|
|
VkStructureType sType;
|
|
|
|
void* pNext;
|
|
|
|
VkBool32 fullScreenExclusiveSupported;
|
|
|
|
} VkSurfaceCapabilitiesFullScreenExclusiveEXT;
|
|
|
|
|
|
|
|
typedef struct VkSurfaceFullScreenExclusiveWin32InfoEXT {
|
|
|
|
VkStructureType sType;
|
|
|
|
const void* pNext;
|
|
|
|
HMONITOR hmonitor;
|
|
|
|
} VkSurfaceFullScreenExclusiveWin32InfoEXT;
|
|
|
|
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetPhysicalDeviceSurfacePresentModes2EXT)(VkPhysicalDevice physicalDevice, const VkPhysicalDeviceSurfaceInfo2KHR* pSurfaceInfo, uint32_t* pPresentModeCount, VkPresentModeKHR* pPresentModes);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkAcquireFullScreenExclusiveModeEXT)(VkDevice device, VkSwapchainKHR swapchain);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkReleaseFullScreenExclusiveModeEXT)(VkDevice device, VkSwapchainKHR swapchain);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetDeviceGroupSurfacePresentModes2EXT)(VkDevice device, const VkPhysicalDeviceSurfaceInfo2KHR* pSurfaceInfo, VkDeviceGroupPresentModeFlagsKHR* pModes);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetPhysicalDeviceSurfacePresentModes2EXT(
|
|
|
|
VkPhysicalDevice physicalDevice,
|
|
|
|
const VkPhysicalDeviceSurfaceInfo2KHR* pSurfaceInfo,
|
|
|
|
uint32_t* pPresentModeCount,
|
|
|
|
VkPresentModeKHR* pPresentModes);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkAcquireFullScreenExclusiveModeEXT(
|
|
|
|
VkDevice device,
|
|
|
|
VkSwapchainKHR swapchain);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkReleaseFullScreenExclusiveModeEXT(
|
|
|
|
VkDevice device,
|
|
|
|
VkSwapchainKHR swapchain);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetDeviceGroupSurfacePresentModes2EXT(
|
|
|
|
VkDevice device,
|
|
|
|
const VkPhysicalDeviceSurfaceInfo2KHR* pSurfaceInfo,
|
|
|
|
VkDeviceGroupPresentModeFlagsKHR* pModes);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
|
|
#define VK_NV_acquire_winrt_display 1
|
|
|
|
#define VK_NV_ACQUIRE_WINRT_DISPLAY_SPEC_VERSION 1
|
|
|
|
#define VK_NV_ACQUIRE_WINRT_DISPLAY_EXTENSION_NAME "VK_NV_acquire_winrt_display"
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkAcquireWinrtDisplayNV)(VkPhysicalDevice physicalDevice, VkDisplayKHR display);
|
|
|
|
typedef VkResult (VKAPI_PTR *PFN_vkGetWinrtDisplayNV)(VkPhysicalDevice physicalDevice, uint32_t deviceRelativeId, VkDisplayKHR* pDisplay);
|
|
|
|
|
|
|
|
#ifndef VK_NO_PROTOTYPES
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkAcquireWinrtDisplayNV(
|
|
|
|
VkPhysicalDevice physicalDevice,
|
|
|
|
VkDisplayKHR display);
|
|
|
|
|
|
|
|
VKAPI_ATTR VkResult VKAPI_CALL vkGetWinrtDisplayNV(
|
|
|
|
VkPhysicalDevice physicalDevice,
|
|
|
|
uint32_t deviceRelativeId,
|
|
|
|
VkDisplayKHR* pDisplay);
|
|
|
|
#endif
|
|
|
|
|
Merge pull request #12703 from wzw-intel:vkcom
* dnn: Add a Vulkan based backend
This commit adds a new backend "DNN_BACKEND_VKCOM" and a
new target "DNN_TARGET_VULKAN". VKCOM means vulkan based
computation library.
This backend uses Vulkan API and SPIR-V shaders to do
the inference computation for layers. The layer types
that implemented in DNN_BACKEND_VKCOM include:
Conv, Concat, ReLU, LRN, PriorBox, Softmax, MaxPooling,
AvePooling, Permute
This is just a beginning work for Vulkan in OpenCV DNN,
more layer types will be supported and performance
tuning is on the way.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/vulkan: Add FindVulkan.cmake to detect Vulkan SDK
In order to build dnn with Vulkan support, need installing
Vulkan SDK and setting environment variable "VULKAN_SDK" and
add "-DWITH_VULKAN=ON" to cmake command.
You can download Vulkan SDK from:
https://vulkan.lunarg.com/sdk/home#linux
For how to install, see
https://vulkan.lunarg.com/doc/sdk/latest/linux/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/windows/getting_started.html
https://vulkan.lunarg.com/doc/sdk/latest/mac/getting_started.html
respectively for linux, windows and mac.
To run the vulkan backend, also need installing mesa driver.
On Ubuntu, use this command 'sudo apt-get install mesa-vulkan-drivers'
To test, use command '$BUILD_DIR/bin/opencv_test_dnn --gtest_filter=*VkCom*'
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: dynamically load Vulkan runtime
No compile-time dependency on Vulkan library.
If Vulkan runtime is unavailable, fallback to CPU path.
Use environment "OPENCL_VULKAN_RUNTIME" to specify path to your
own vulkan runtime library.
Signed-off-by: Wu Zhiwen <zhiwen.wu@intel.com>
* dnn/Vulkan: Add a python script to compile GLSL shaders to SPIR-V shaders
The SPIR-V shaders are in format of text-based 32-bit hexadecimal
numbers, and inserted into .cpp files as unsigned int32 array.
* dnn/Vulkan: Put Vulkan headers into 3rdparty directory and some other fixes
Vulkan header files are copied from
https://github.com/KhronosGroup/Vulkan-Docs/tree/master/include/vulkan
to 3rdparty/include
Fix the Copyright declaration issue.
Refine OpenCVDetectVulkan.cmake
* dnn/Vulkan: Add vulkan backend tests into existing ones.
Also fixed some test failures.
- Don't use bool variable as uniform for shader
- Fix dispathed group number beyond max issue
- Bypass "group > 1" convolution. This should be support in future.
* dnn/Vulkan: Fix multiple initialization in one thread.
6 years ago
|
|
|
#ifdef __cplusplus
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#endif
|