Merge branch 'dev/alexey/master' into dev/cenit/deprecate_msbuild

AlexeyAB · Aug 27, 2023 · f8e3685 · f8e3685
2 parents 6f975a3 + 538c435
commit f8e3685
Show file tree

Hide file tree

Showing 33 changed files with 616 additions and 239 deletions.
diff --git a/.github/workflows/ccpp.yml b/.github/workflows/ccpp.yml
@@ -24,7 +24,7 @@ jobs:
     - name: Update apt
       run: sudo apt update
     - name: Install dependencies
-      run: sudo apt install libopencv-dev
+      run: sudo apt install libopencv-dev libgles2-mesa-dev
 
     - name: 'Install CUDA'
       run: ${{ github.workspace }}/scripts/deploy-cuda.sh
@@ -293,7 +293,7 @@ jobs:
     - name: Update apt
       run: sudo apt update
     - name: Install dependencies
-      run: sudo apt install libopencv-dev
+      run: sudo apt install libopencv-dev libgles2-mesa-dev
 
     - uses: lukka/get-cmake@latest
 
@@ -656,9 +656,9 @@ jobs:
 
     - name: 'Build'
       env:
-        CUDA_PATH: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.7"
-        CUDA_TOOLKIT_ROOT_DIR: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.7"
-        CUDACXX: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.7\\bin\\nvcc.exe"
+        CUDA_PATH: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.2"
+        CUDA_TOOLKIT_ROOT_DIR: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.2"
+        CUDACXX: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.2\\bin\\nvcc.exe"
       shell: pwsh
       run: ${{ github.workspace }}/build.ps1 -EnableCUDA -DisableInteractive -DoNotUpdateTOOL
 

diff --git a/.github/workflows/on_pr.yml b/.github/workflows/on_pr.yml
@@ -14,7 +14,7 @@ jobs:
     - name: Update apt
       run: sudo apt update
     - name: Install dependencies
-      run: sudo apt install libopencv-dev
+      run: sudo apt install libopencv-dev libgles2-mesa-dev
 
     - name: 'Install CUDA'
       run: ${{ github.workspace }}/scripts/deploy-cuda.sh
@@ -215,7 +215,7 @@ jobs:
     - name: Update apt
       run: sudo apt update
     - name: Install dependencies
-      run: sudo apt install libopencv-dev
+      run: sudo apt install libopencv-dev libgles2-mesa-dev
 
     - uses: lukka/get-cmake@latest
 
@@ -408,9 +408,9 @@ jobs:
 
     - name: 'Build'
       env:
-        CUDA_PATH: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.7"
-        CUDA_TOOLKIT_ROOT_DIR: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.7"
-        CUDACXX: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.7\\bin\\nvcc.exe"
+        CUDA_PATH: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.2"
+        CUDA_TOOLKIT_ROOT_DIR: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.2"
+        CUDACXX: "C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v12.2\\bin\\nvcc.exe"
       shell: pwsh
       run: ${{ github.workspace }}/build.ps1 -EnableCUDA -DisableInteractive -DoNotUpdateTOOL
 

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -9,6 +9,7 @@ option(BUILD_USELIB_TRACK "Build uselib_track" ON)
 option(MANUALLY_EXPORT_TRACK_OPTFLOW "Manually export the TRACK_OPTFLOW=1 define" OFF)
 option(ENABLE_OPENCV "Enable OpenCV integration" ON)
 option(ENABLE_CUDA "Enable CUDA support" ON)
+cmake_dependent_option(ENABLE_CUDA_OPENGL_INTEGRATION "Build darknet with support for running networks straight from OpenGL textures" ON "ENABLE_CUDA" OFF)
 option(ENABLE_CUDNN "Enable CUDNN" ON)
 option(ENABLE_CUDNN_HALF "Enable CUDNN Half precision" ON)
 option(ENABLE_ZED_CAMERA "Enable ZED Camera support" ON)
@@ -74,8 +75,6 @@ else()
   set(IS_X86 FALSE)
 endif()
 
-cmake_dependent_option(ENABLE_SSE_AND_AVX_FLAGS "Enable AVX and SSE optimizations (x86-only)" ON "CMAKE_COMPILER_IS_GNUCC_OR_CLANG;IS_X86" OFF)
-
 if(ENABLE_VCPKG_INTEGRATION AND DEFINED ENV{VCPKG_ROOT} AND NOT DEFINED CMAKE_TOOLCHAIN_FILE)
   set(CMAKE_POLICY_DEFAULT_CMP0077 NEW)
   set(X_VCPKG_APPLOCAL_DEPS_INSTALL ON)
@@ -150,6 +149,8 @@ else()
   set(CMAKE_COMPILER_IS_CLANG FALSE)
 endif()
 
+cmake_dependent_option(ENABLE_SSE_AND_AVX_FLAGS "Enable AVX and SSE optimizations (x86-only)" ON "CMAKE_COMPILER_IS_GNUCC_OR_CLANG;IS_X86" OFF)
+
 set(default_build_type "Release")
 if(NOT CMAKE_BUILD_TYPE AND NOT CMAKE_CONFIGURATION_TYPES)
   message(STATUS "Setting build type to '${default_build_type}' as none was specified.")
@@ -327,6 +328,11 @@ endif()
 if(ENABLE_CUDA)
   if(MSVC)
     set(ADDITIONAL_CXX_FLAGS "${ADDITIONAL_CXX_FLAGS} /DGPU")
+
+    if(ENABLE_CUDA_OPENGL_INTEGRATION)
+      set(ADDITIONAL_CXX_FLAGS "${ADDITIONAL_CXX_FLAGS} /DCUDA_OPENGL_INTEGRATION")
+    endif()
+
     if(CUDNN_FOUND)
       set(ADDITIONAL_CXX_FLAGS "${ADDITIONAL_CXX_FLAGS} /DCUDNN")
     endif()
@@ -337,6 +343,11 @@ if(ENABLE_CUDA)
     set(CUDA_HOST_COMPILER_FLAGS "-Wno-deprecated-declarations -Xcompiler=\"${ADDITIONAL_CXX_FLAGS_COMMA_SEPARATED}\"")
   else()
     set(ADDITIONAL_CXX_FLAGS "${ADDITIONAL_CXX_FLAGS} -DGPU")
+
+    if(ENABLE_CUDA_OPENGL_INTEGRATION)
+      set(ADDITIONAL_CXX_FLAGS "${ADDITIONAL_CXX_FLAGS} -DCUDA_OPENGL_INTEGRATION")
+    endif()
+
     if(CUDNN_FOUND)
       set(ADDITIONAL_CXX_FLAGS "${ADDITIONAL_CXX_FLAGS} -DCUDNN")
     endif()
@@ -544,6 +555,11 @@ if(ENABLE_CUDA)
   target_compile_definitions(dark PUBLIC -DGPU)
 endif()
 
+if(ENABLE_CUDA_OPENGL_INTEGRATION)
+  target_compile_definitions(darknet PRIVATE -DCUDA_OPENGL_INTEGRATION)
+  target_compile_definitions(dark PUBLIC -DCUDA_OPENGL_INTEGRATION)
+endif()
+
 if(USE_INTEGRATED_LIBS AND WIN32)
   target_compile_definitions(darknet PRIVATE -D_TIMESPEC_DEFINED)
   target_compile_definitions(dark PRIVATE -D_TIMESPEC_DEFINED)

diff --git a/Dockerfile.cpu b/Dockerfile.cpu
@@ -0,0 +1,49 @@
+FROM ubuntu:latest AS builder
+
+ENV DEBIAN_FRONTEND noninteractive
+
+RUN apt-get update -y
+
+RUN apt-get install -y g++ make pkg-config libopencv-dev
+
+COPY  . /darknet
+
+WORKDIR /darknet
+
+RUN rm Dockerfile.cpu
+
+RUN rm Dockerfile.gpu
+
+RUN rm Docker-compose.yml
+
+RUN make
+
+FROM ubuntu:latest
+
+ENV DEBIAN_FRONTEND noninteractive
+
+RUN apt-get update -y
+
+RUN apt-get install -y sudo libgomp1
+
+RUN useradd -U -m yolo
+
+RUN usermod -aG sudo yolo 
+
+RUN usermod --shell /bin/bash yolo
+
+RUN echo "yolo:yolo" | chpasswd
+
+COPY --from=builder /darknet /home/yolo/darknet
+
+RUN cp /home/yolo/darknet/libdarknet.so /usr/local/lib/libdarknet.so || echo "libso not used"
+
+RUN cp /home/yolo/darknet/include/darknet.h /usr/local/include/darknet.h
+
+RUN ldconfig
+
+WORKDIR /home/yolo/darknet
+
+USER yolo
+
+
diff --git a/Dockerfile.gpu b/Dockerfile.gpu
@@ -0,0 +1,47 @@
+FROM nvidia/cuda:11.6.0-cudnn8-devel-ubuntu20.04 AS builder
+
+ENV DEBIAN_FRONTEND noninteractive
+
+RUN apt-get update -y
+
+RUN apt-get install -y g++ make pkg-config libopencv-dev
+
+COPY  . /darknet
+
+WORKDIR /darknet
+
+RUN rm Dockerfile.cpu
+
+RUN rm Dockerfile.gpu
+
+RUN rm Docker-compose.yml
+
+RUN make
+
+FROM nvidia/cuda:11.6.0-cudnn8-devel-ubuntu20.04
+
+ENV DEBIAN_FRONTEND noninteractive
+
+RUN apt-get update -y
+
+RUN apt-get install -y sudo libgomp1
+
+RUN useradd -U -m yolo
+
+RUN usermod -aG sudo yolo 
+
+RUN usermod --shell /bin/bash yolo
+
+RUN echo "yolo:yolo" | chpasswd
+
+COPY --from=builder /darknet /home/yolo/darknet
+
+RUN cp /home/yolo/darknet/libdarknet.so /usr/local/lib/libdarknet.so || echo "libso not used"
+
+RUN cp /home/yolo/darknet/include/darknet.h /usr/local/include/darknet.h
+
+RUN ldconfig
+
+WORKDIR /home/yolo/darknet
+
+USER yolo
diff --git a/Makefile b/Makefile
@@ -9,54 +9,70 @@ ZED_CAMERA=0
 ZED_CAMERA_v2_8=0
 
 # set GPU=1 and CUDNN=1 to speedup on GPU
-# set CUDNN_HALF=1 to further speedup 3 x times (Mixed-precision on Tensor Cores) GPU: Volta, Xavier, Turing and higher
+# set CUDNN_HALF=1 to further speedup 3 x times (Mixed-precision on Tensor Cores) GPU: Volta, Xavier, Turing, Ampere, Ada and higher
 # set AVX=1 and OPENMP=1 to speedup on CPU (if error occurs then set AVX=0)
 # set ZED_CAMERA=1 to enable ZED SDK 3.0 and above
 # set ZED_CAMERA_v2_8=1 to enable ZED SDK 2.X
 
 USE_CPP=0
 DEBUG=0
 
-ARCH= -gencode arch=compute_35,code=sm_35 \
-      -gencode arch=compute_50,code=[sm_50,compute_50] \
+ARCH= -gencode arch=compute_50,code=[sm_50,compute_50] \
       -gencode arch=compute_52,code=[sm_52,compute_52] \
 	    -gencode arch=compute_61,code=[sm_61,compute_61]
 
 OS := $(shell uname)
 
-# GeForce RTX 3070, 3080, 3090
-# ARCH= -gencode arch=compute_86,code=[sm_86,compute_86]
+# Naming confusion with recent RTX cards.
+# "NVIDIA Quadro RTX x000" and T1000/Tx00 are Turing Architecture Family with Compute Capability of 7.5
+# "NVIDIA RTX Ax000" are Ampere Architecture Family with Compute Capability of 8.6
+# NVIDIA "RTX x000 Ada" are Ada Lovelace Architecture Family with Compute Capability of 8.9
+# Source https://developer.nvidia.com/cuda-gpus
 
-# Kepler GeForce GTX 770, GTX 760, GT 740
+# KEPLER, GeForce GTX 770, GTX 760, GT 740
 # ARCH= -gencode arch=compute_30,code=sm_30
 
-# Tesla A100 (GA100), DGX-A100, RTX 3080
-# ARCH= -gencode arch=compute_80,code=[sm_80,compute_80]
+# MAXWELL, GeForce GTX 950, 960, 970, 980, 980 Ti, "GTX" Titan X
+# ARCH= -gencode arch=compute_52,code=sm_52
+
+# Jetson TX1, Tegra X1, DRIVE CX, DRIVE PX, Jetson Nano (2GB, 4GB)
+# ARCH= -gencode arch=compute_53,code=[sm_53,compute_53]
+
+# GP100/Tesla P100 - DGX-1
+# ARCH= -gencode arch=compute_60,code=sm_60
+
+# PASCAL, GTX 10x0, GTX 10x0 Ti, Titan Xp, Tesla P40, Tesla P4
+# ARCH= -gencode arch=compute_61,code=[sm_61,compute_61]
+
+# For Jetson TX2, Jetson Nano TX2 or Drive-PX2 uncomment:
+# ARCH= -gencode arch=compute_62,code=[sm_62,compute_62]
 
 # Tesla V100
 # ARCH= -gencode arch=compute_70,code=[sm_70,compute_70]
 
-# GeForce RTX 2080 Ti, RTX 2080, RTX 2070, Quadro RTX 8000, Quadro RTX 6000, Quadro RTX 5000, Tesla T4, XNOR Tensor Cores
-# ARCH= -gencode arch=compute_75,code=[sm_75,compute_75]
-
-# Jetson XAVIER
+# Jetson XAVIER, XAVIER NX
 # ARCH= -gencode arch=compute_72,code=[sm_72,compute_72]
 
-# GTX 1080, GTX 1070, GTX 1060, GTX 1050, GTX 1030, Titan Xp, Tesla P40, Tesla P4
-# ARCH= -gencode arch=compute_61,code=sm_61 -gencode arch=compute_61,code=compute_61
+# GeForce Titan RTX, RTX 20x0, RTX 20x0 Ti, Quadro RTX x000, Tesla T4, XNOR Tensor Cores
+# ARCH= -gencode arch=compute_75,code=[sm_75,compute_75]
 
-# GP100/Tesla P100 - DGX-1
-# ARCH= -gencode arch=compute_60,code=sm_60
+# Tesla A100 (GA100), DGX-A100, A30, A100, RTX 3080
+# ARCH= -gencode arch=compute_80,code=[sm_80,compute_80]
 
-# For Jetson TX1, Tegra X1, DRIVE CX, DRIVE PX - uncomment:
-# ARCH= -gencode arch=compute_53,code=[sm_53,compute_53]
+# GeForce RTX 30x0, 30x0 Ti, Tesla GA10x, RTX Axxxx, A2, A10, A16, A40
+# ARCH= -gencode arch=compute_86,code=[sm_86,compute_86]
 
-# For Jetson Tx2 or Drive-PX2 uncomment:
-# ARCH= -gencode arch=compute_62,code=[sm_62,compute_62]
+# NOT TESTED, THEORETICAL
+# Jetson ORIN, ORIN NX, ORIN NANO
+# ARCH= -gencode arch=compute_87,code=[sm_87,compute_87]
 
-# For Tesla GA10x cards, RTX 3090, RTX 3080, RTX 3070, RTX A6000, RTX A40 uncomment:
-# ARCH= -gencode arch=compute_86,code=[sm_86,compute_86]
+# NOT TESTED, THEORETICAL
+# GeForce RTX 4070 Ti, 4080, 4090, L4, L40
+# ARCH= -gencode arch=compute_89,code=[sm_89,compute_89]
 
+# NOT TESTED, THEORETICAL
+# Nvidia H100
+# ARCH= -gencode arch=compute_90,code=[sm_90,compute_90]
 
 VPATH=./src/
 EXEC=darknet