 [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago  [WIP] vulkan compute (#618)
* vulkan infrastructure
* vkallocator and vkmat
* layer interface for vulkan compute
* wip...
* default vulkan device, command wrapper, upload model weight in load_model to simplify layer interface
* simplify command api, vkmat holds staging buffer, relu works
* initialize specialization constant, simplify command dispatch, fix staging buffer copy with different shape, convolution works
* init extension functions
* dynamic local size and group count
* group count=1 is invalid
* regard device max workgroup size limit
* fix relu oooops
* decouple command record and staging allocation
* create result blob
* add pooling shader
* buffer is faster than image :)
* fix pooling shader
* add innerproduct shader
* readonly writeonly decoration
* simplify buffer creation
* decouple command and layer, VK_KHR_descriptor_update_template extension makes descriptor binding update easy :D
* fix vulkan building issues in visual studio (#1)
* fix building issues on visual studio
* ignore benchmark
* cancel changes
* ... ...
* decouple paramdict and vulkandevice
* fix staging buffer destroy in model loading
* remove vkdev member in option
* add padding shader
* simplify vulkan layer creation, simplify convolution and pooling shader for no padding, less debug output
* add convolutiondepthwise and softmax shader
* specialization float type, add leakyrelu
* add dropout shader
* add batchnorm shader
* split vulkan forward
* add scale shader
* push constant type can be int or float
* set_optimal_local_size_xyz
* add eltwise shader
* concat vulkan forward
* fix convolution without bias
* add dummy shader for concat and split, more fix ...
* optional VK_KHR_descriptor_update_template and VK_KHR_push_descriptor
* check VK_KHR_push_descriptor for vkCmdPushDescriptorSetWithTemplateKHR
* binaryop and unaryop shader
* hide raw command buffer
* simple vkbenchncnn benchmark
* create device with transfer queue
* rename command to vkcompute, add vktransfer and layer upload_model interface
* external VkMat, copy and map wrt buffer offset
* command copy respect offset and size
* decouple weight upload and load, simplify upload weight api, use one big staging buffer for uploading weights
* fix build on android
* binding count can not vary :(
* barrier check state, fix sub-op destruction
* declare local_size_xyz constant, fix crash on radv
* fix local_size_xyz, second try
* more barrier and state fix
* fix softmax
* reconstruct buffer memory allocator, reuse blob buffer, less verbose output
* find unified memory type index
* weight staging buffer allocator and weight buffer allocator, respect descriptor buffer offset alignment
* use VK_KHR_descriptor_update_template for faster descriptor update if available, multithread pipeline creation
* find more useful vulkan extensions and enable them
* fix msvc build
* respect VK_KHR_dedicated_allocation for weight buffer allocation
* fix android build
* fix bias name conflicts with metal
* decouple pipeline and layer, building shader sources into shader module, dedicated create_pipeline api, simplify pipeline recording
* drop dummy shader, inplace softmax, multiple shader module works
* fix unique queue family index error
* flatten support vulkan
* mnasnet run
* find shader module by name, each entry point per shader module, fix attribute/id conflict on moltenvk
* some minor changes
* add some high level api
* use dedicated transfer queue to upload weight model
* prefer mappable buffer on unified memory
* global pooling and convolution fc, reuse staging buffer
* implement ring-buffer style blob allocator, add VkBufferMemory capacity
* use blob allocator for workspace blob, it works fine :)
* vulkan option off
* Update layer.cpp
* fix build with vulkan off
* less verbose output, fix crash on vulkan_compute off
* merge benchncnn tool
* allocator clear api, use new weight buffer allocator per net
* add default locked allocator
* mapped mat ptr api, persistent mapped memory works generally :)
* travis ci linux vulkan
* travis ci vulkan wip ...
* more gpu wip ...
* more gpu wip ...
* wip...
* wip...
* wip... ...
* wip... ios vulkan build...
* find glslangValidator on ios build
* use dynamic moltenvk library
* travis ci wip ...
* ios simulator does not support metal at all
* fix cpu only extractor
* optimize workgroup size, first try
* optimize workgroup size, second try
* conv1x1s1d1 vec4
* revert build system
* fix ncnn2mem build
* fix ncnn2mem build
7 years ago |
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467 |
- // Tencent is pleased to support the open source community by making ncnn available.
- //
- // Copyright (C) 2019 THL A29 Limited, a Tencent company. All rights reserved.
- //
- // Licensed under the BSD 3-Clause License (the "License"); you may not use this file except
- // in compliance with the License. You may obtain a copy of the License at
- //
- // https://opensource.org/licenses/BSD-3-Clause
- //
- // 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.
-
- #include "pipeline.h"
-
- #include "layer_shader_type.h"
- #include "mat.h"
- #include "pipelinecache.h"
- #include "option.h"
-
- #include <math.h>
-
- #if __ANDROID_API__ >= 26
- #include <android/hardware_buffer.h>
- #endif // __ANDROID_API__ >= 26
-
- namespace ncnn {
-
- #if NCNN_VULKAN
- class PipelinePrivate
- {
- public:
- VkShaderModule shader_module;
- VkDescriptorSetLayout descriptorset_layout;
- VkPipelineLayout pipeline_layout;
- VkPipeline pipeline;
- VkDescriptorUpdateTemplateKHR descriptor_update_template;
-
- ShaderInfo shader_info;
-
- uint32_t local_size_x;
- uint32_t local_size_y;
- uint32_t local_size_z;
- };
-
- Pipeline::Pipeline(const VulkanDevice* _vkdev)
- : vkdev(_vkdev), d(new PipelinePrivate)
- {
- d->shader_module = 0;
- d->descriptorset_layout = 0;
- d->pipeline_layout = 0;
- d->pipeline = 0;
- d->descriptor_update_template = 0;
-
- d->local_size_x = 1;
- d->local_size_y = 1;
- d->local_size_z = 1;
- }
-
- Pipeline::~Pipeline()
- {
- delete d;
- }
-
- Pipeline::Pipeline(const Pipeline&)
- : d(0)
- {
- }
-
- Pipeline& Pipeline::operator=(const Pipeline&)
- {
- return *this;
- }
-
- void Pipeline::set_optimal_local_size_xyz(int w, int h, int c)
- {
- set_optimal_local_size_xyz(Mat(w, h, c, (void*)0));
- }
-
- void Pipeline::set_optimal_local_size_xyz(const Mat& local_size_xyz)
- {
- int w = local_size_xyz.w;
- int h = local_size_xyz.h;
- int c = local_size_xyz.c;
-
- if (w == 0 && h == 0 && c == 0)
- {
- // fallback to the common and safe 4x4x4
- w = 4;
- h = 4;
- c = 4;
- }
-
- w = std::min(w, (int)vkdev->info.max_workgroup_size_x());
- h = std::min(h, (int)vkdev->info.max_workgroup_size_y());
- c = std::min(c, (int)vkdev->info.max_workgroup_size_z());
-
- if (w * h * c <= (int)vkdev->info.max_workgroup_invocations())
- {
- return set_local_size_xyz(w, h, c);
- }
-
- int max_local_size_xy = (int)vkdev->info.max_workgroup_invocations() / c;
-
- int wh_max = std::max(1, (int)sqrt(max_local_size_xy));
- while (w * h >= wh_max)
- {
- w = std::max(1, w / 2);
- h = std::max(1, h / 2);
- }
-
- set_local_size_xyz(w, h, c);
- }
-
- void Pipeline::set_local_size_xyz(int w, int h, int c)
- {
- d->local_size_x = w;
- d->local_size_y = h;
- d->local_size_z = c;
-
- // NCNN_LOGE("local size = %d %d %d", local_size_x, local_size_y, local_size_z);
- }
-
- int Pipeline::create(const uint32_t* spv_data, size_t spv_data_size, const std::vector<vk_specialization_type>& specializations)
- {
- const PipelineCache* pipeline_cache = vkdev->get_pipeline_cache();
-
- // get from pipeline cache
- return pipeline_cache->get_pipeline(spv_data, spv_data_size, specializations, d->local_size_x, d->local_size_y, d->local_size_z,
- &d->shader_module, &d->descriptorset_layout, &d->pipeline_layout, &d->pipeline, &d->descriptor_update_template,
- d->shader_info);
- }
-
- int Pipeline::create(int shader_type_index, const Option& opt, const std::vector<vk_specialization_type>& specializations)
- {
- const PipelineCache* pipeline_cache = opt.pipeline_cache ? opt.pipeline_cache : vkdev->get_pipeline_cache();
-
- // get from pipeline cache
- return pipeline_cache->get_pipeline(shader_type_index, opt, specializations, d->local_size_x, d->local_size_y, d->local_size_z,
- &d->shader_module, &d->descriptorset_layout, &d->pipeline_layout, &d->pipeline, &d->descriptor_update_template,
- d->shader_info);
- }
-
- VkShaderModule Pipeline::shader_module() const
- {
- return d->shader_module;
- }
-
- VkDescriptorSetLayout Pipeline::descriptorset_layout() const
- {
- return d->descriptorset_layout;
- }
-
- VkPipelineLayout Pipeline::pipeline_layout() const
- {
- return d->pipeline_layout;
- }
-
- VkPipeline Pipeline::pipeline() const
- {
- return d->pipeline;
- }
-
- VkDescriptorUpdateTemplateKHR Pipeline::descriptor_update_template() const
- {
- return d->descriptor_update_template;
- }
-
- const ShaderInfo& Pipeline::shader_info() const
- {
- return d->shader_info;
- }
-
- uint32_t Pipeline::local_size_x() const
- {
- return d->local_size_x;
- }
-
- uint32_t Pipeline::local_size_y() const
- {
- return d->local_size_y;
- }
-
- uint32_t Pipeline::local_size_z() const
- {
- return d->local_size_z;
- }
-
- void Pipeline::set_shader_module(VkShaderModule shader_module)
- {
- d->shader_module = shader_module;
- }
-
- void Pipeline::set_descriptorset_layout(VkDescriptorSetLayout descriptorset_layout)
- {
- d->descriptorset_layout = descriptorset_layout;
- }
-
- void Pipeline::set_pipeline_layout(VkPipelineLayout pipeline_layout)
- {
- d->pipeline_layout = pipeline_layout;
- }
-
- void Pipeline::set_pipeline(VkPipeline pipeline)
- {
- d->pipeline = pipeline;
- }
-
- void Pipeline::set_descriptor_update_template(VkDescriptorUpdateTemplateKHR descriptor_update_template)
- {
- d->descriptor_update_template = descriptor_update_template;
- }
-
- void Pipeline::set_shader_info(const ShaderInfo& shader_info)
- {
- d->shader_info = shader_info;
- }
-
- #if NCNN_PLATFORM_API
- #if __ANDROID_API__ >= 26
- ImportAndroidHardwareBufferPipeline::ImportAndroidHardwareBufferPipeline(const VulkanDevice* _vkdev)
- : Pipeline(_vkdev)
- {
- sampler = 0;
- }
-
- ImportAndroidHardwareBufferPipeline::~ImportAndroidHardwareBufferPipeline()
- {
- destroy();
- }
-
- int ImportAndroidHardwareBufferPipeline::create(VkAndroidHardwareBufferImageAllocator* ahb_im_allocator, int _type_to, int _rotate_from, const Option& opt)
- {
- int target_width;
- int target_height;
-
- if (rotate_from < 5) // 1 2 3 4
- {
- target_width = ahb_im_allocator->width();
- target_height = ahb_im_allocator->height();
- }
- else // 5 6 7 8
- {
- target_width = ahb_im_allocator->height();
- target_height = ahb_im_allocator->width();
- }
-
- return create(ahb_im_allocator, _type_to, _rotate_from, target_width, target_height, opt);
- }
-
- int ImportAndroidHardwareBufferPipeline::create(VkAndroidHardwareBufferImageAllocator* ahb_im_allocator, int _type_to, int _rotate_from, int target_width, int target_height, const Option& opt)
- {
- int w = ahb_im_allocator->width();
- int h = ahb_im_allocator->height();
-
- type_to = _type_to;
- rotate_from = _rotate_from;
-
- need_resize = false;
- if (rotate_from < 5) // 1 2 3 4
- {
- if (target_width != w || target_height != h)
- need_resize = true;
- }
- else // 5 6 7 8
- {
- if (target_width != h || target_height != w)
- need_resize = true;
- }
-
- // if (type_to == 1 || type_to == 2)
- // {
- // outc = 3;
- // out_elemsize = vkdev->info.support_fp16_storage() && opt.use_fp16_storage ? 2u : 4u;
- // out_elempack = 1;
- // }
- // else if (type_to == 3)
- // {
- // outc = 1;
- // out_elemsize = vkdev->info.support_fp16_storage() && opt.use_fp16_storage ? 2u : 4u;
- // out_elempack = 1;
- // }
- // else // if (type_to == 4 || type_to == 5)
- // {
- // outc = 1;
- // out_elemsize = ((vkdev->info.support_fp16_packed() && opt.use_fp16_packed) || (vkdev->info.support_fp16_storage() && opt.use_fp16_storage)) ? 8u : 16u;
- // out_elempack = 4;
- // }
-
- set_local_size_xyz(8, 8, 1);
-
- std::vector<vk_specialization_type> specializations(7);
- specializations[0].i = ahb_im_allocator->width();
- specializations[1].i = ahb_im_allocator->height();
- specializations[2].i = target_width;
- specializations[3].i = target_height;
- specializations[4].i = type_to;
- specializations[5].i = rotate_from;
- specializations[6].i = need_resize;
-
- create_shader_module(opt);
-
- const ShaderInfo& _shader_info = shader_info();
-
- if ((int)specializations.size() != _shader_info.specialization_count)
- {
- NCNN_LOGE("pipeline convert_ycbcr specialization count mismatch, expect %d but got %d", _shader_info.specialization_count, (int)specializations.size());
- return -1;
- }
-
- create_sampler(ahb_im_allocator);
-
- create_descriptorset_layout();
-
- VkPipelineLayout pipeline_layout = 0;
- VkPipeline pipeline = 0;
- VkDescriptorUpdateTemplateKHR descriptor_update_template = 0;
-
- vkdev->create_pipeline_layout(_shader_info.push_constant_count, descriptorset_layout(), &pipeline_layout);
-
- vkdev->create_pipeline(shader_module(), pipeline_layout, specializations, &pipeline);
-
- if (vkdev->info.support_VK_KHR_descriptor_update_template())
- {
- vkdev->create_descriptor_update_template(_shader_info.binding_count, _shader_info.binding_types, descriptorset_layout(), pipeline_layout, &descriptor_update_template);
- }
-
- set_pipeline_layout(pipeline_layout);
- set_pipeline(pipeline);
- set_descriptor_update_template(descriptor_update_template);
-
- return 0;
- }
-
- void ImportAndroidHardwareBufferPipeline::destroy()
- {
- if (sampler)
- {
- vkDestroySampler(vkdev->vkdevice(), sampler, 0);
- sampler = 0;
- }
- }
-
- int ImportAndroidHardwareBufferPipeline::create_shader_module(const Option& opt)
- {
- int shader_type_index = LayerShaderType::convert_ycbcr;
-
- std::vector<uint32_t> spirv;
- int retc = compile_spirv_module(shader_type_index, opt, spirv);
- if (retc != 0)
- {
- NCNN_LOGE("compile_spirv_module failed %d", retc);
- return -1;
- }
-
- const uint32_t* spv_data = spirv.data();
- size_t spv_data_size = spirv.size() * 4;
-
- ShaderInfo shader_info;
- int ret = resolve_shader_info(spv_data, spv_data_size, shader_info);
- if (ret != 0)
- {
- NCNN_LOGE("resolve_shader_info failed %d", ret);
- return -1;
- }
-
- set_shader_info(shader_info);
-
- VkShaderModule shader_module = vkdev->compile_shader_module(spv_data, spv_data_size, local_size_x(), local_size_y(), local_size_z());
- set_shader_module(shader_module);
-
- return 0;
- }
-
- int ImportAndroidHardwareBufferPipeline::create_sampler(VkAndroidHardwareBufferImageAllocator* ahb_im_allocator)
- {
- VkResult ret;
-
- VkExternalFormatANDROID externalFormatANDROID;
- externalFormatANDROID.sType = VK_STRUCTURE_TYPE_EXTERNAL_FORMAT_ANDROID;
- externalFormatANDROID.pNext = 0;
- externalFormatANDROID.externalFormat = ahb_im_allocator->external_format();
-
- VkSamplerYcbcrConversionInfoKHR samplerYcbcrConversionInfo;
- samplerYcbcrConversionInfo.sType = VK_STRUCTURE_TYPE_SAMPLER_YCBCR_CONVERSION_INFO_KHR;
- samplerYcbcrConversionInfo.pNext = &externalFormatANDROID;
- samplerYcbcrConversionInfo.conversion = ahb_im_allocator->samplerYcbcrConversion;
-
- VkSamplerCreateInfo samplerCreateInfo;
- samplerCreateInfo.sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO;
- samplerCreateInfo.pNext = &samplerYcbcrConversionInfo;
- samplerCreateInfo.magFilter = need_resize ? VK_FILTER_LINEAR : VK_FILTER_NEAREST;
- samplerCreateInfo.minFilter = need_resize ? VK_FILTER_LINEAR : VK_FILTER_NEAREST;
- samplerCreateInfo.mipmapMode = VK_SAMPLER_MIPMAP_MODE_NEAREST;
- samplerCreateInfo.addressModeU = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE;
- samplerCreateInfo.addressModeV = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE;
- samplerCreateInfo.addressModeW = VK_SAMPLER_ADDRESS_MODE_CLAMP_TO_EDGE;
- samplerCreateInfo.mipLodBias = 0.0f;
- samplerCreateInfo.anisotropyEnable = VK_FALSE;
- samplerCreateInfo.maxAnisotropy = 1;
- samplerCreateInfo.compareEnable = VK_FALSE;
- samplerCreateInfo.compareOp = VK_COMPARE_OP_NEVER;
- samplerCreateInfo.minLod = 0.0f;
- samplerCreateInfo.maxLod = 0.0f;
- samplerCreateInfo.borderColor = VK_BORDER_COLOR_FLOAT_TRANSPARENT_BLACK; //VK_BORDER_COLOR_FLOAT_OPAQUE_WHITE; FIXME
- samplerCreateInfo.unnormalizedCoordinates = VK_TRUE; //VK_FALSE; FIXME ?
-
- ret = vkCreateSampler(vkdev->vkdevice(), &samplerCreateInfo, 0, &sampler);
- if (ret != VK_SUCCESS)
- {
- NCNN_LOGE("vkCreateSampler failed %d", ret);
- return -1;
- }
-
- return 0;
- }
-
- int ImportAndroidHardwareBufferPipeline::create_descriptorset_layout()
- {
- VkDescriptorSetLayoutBinding descriptorSetLayoutBindings[3];
- descriptorSetLayoutBindings[0].binding = 0;
- descriptorSetLayoutBindings[0].descriptorType = VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER;
- descriptorSetLayoutBindings[0].descriptorCount = 1;
- descriptorSetLayoutBindings[0].stageFlags = VK_SHADER_STAGE_COMPUTE_BIT;
- descriptorSetLayoutBindings[0].pImmutableSamplers = &sampler;
- descriptorSetLayoutBindings[1].binding = 1;
- descriptorSetLayoutBindings[1].descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER;
- descriptorSetLayoutBindings[1].descriptorCount = 1;
- descriptorSetLayoutBindings[1].stageFlags = VK_SHADER_STAGE_COMPUTE_BIT;
- descriptorSetLayoutBindings[1].pImmutableSamplers = 0;
- descriptorSetLayoutBindings[2].binding = 2;
- descriptorSetLayoutBindings[2].descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER;
- descriptorSetLayoutBindings[2].descriptorCount = 1;
- descriptorSetLayoutBindings[2].stageFlags = VK_SHADER_STAGE_COMPUTE_BIT;
- descriptorSetLayoutBindings[2].pImmutableSamplers = 0;
-
- VkDescriptorSetLayoutCreateInfo descriptorSetLayoutCreateInfo;
- descriptorSetLayoutCreateInfo.sType = VK_STRUCTURE_TYPE_DESCRIPTOR_SET_LAYOUT_CREATE_INFO;
- descriptorSetLayoutCreateInfo.pNext = 0;
- descriptorSetLayoutCreateInfo.flags = 0;
- descriptorSetLayoutCreateInfo.bindingCount = 3;
- descriptorSetLayoutCreateInfo.pBindings = descriptorSetLayoutBindings;
-
- if (vkdev->info.support_VK_KHR_push_descriptor())
- {
- descriptorSetLayoutCreateInfo.flags |= VK_DESCRIPTOR_SET_LAYOUT_CREATE_PUSH_DESCRIPTOR_BIT_KHR;
- }
-
- VkDescriptorSetLayout descriptorset_layout = 0;
- VkResult ret = vkCreateDescriptorSetLayout(vkdev->vkdevice(), &descriptorSetLayoutCreateInfo, 0, &descriptorset_layout);
- if (ret != VK_SUCCESS)
- {
- NCNN_LOGE("vkCreateDescriptorSetLayout failed %d", ret);
- return -1;
- }
-
- set_descriptorset_layout(descriptorset_layout);
-
- return 0;
- }
- #endif // __ANDROID_API__ >= 26
- #endif // NCNN_PLATFORM_API
-
- #endif // NCNN_VULKAN
-
- } // namespace ncnn
|