Blame - docs/00_introduction.dox - ml/ComputeLibrary

blob: 17d1bdea63ea1861b17d7c33df53fc0dedb67bd2 [file] [log] [blame]

Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	1	namespace arm_compute
				2	{
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	3	/** @mainpage Introduction
				4
				5	@tableofcontents
				6
				7	The Computer Vision and Machine Learning library is a set of functions optimised for both ARM CPUs and GPUs using SIMD technologies.
				8
				9	Several builds of the library are available using various configurations:
				10	- OS: Linux, Android or bare metal.
				11	- Architecture: armv7a (32bit) or arm64-v8a (64bit)
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	12	- Technology: NEON / OpenCL / GLES_COMPUTE / NEON and OpenCL and GLES_COMPUTE
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	13	- Debug / Asserts / Release: Use a build with asserts enabled to debug your application and enable extra validation. Once you are sure your application works as expected you can switch to a release build of the library for maximum performance.
				14
				15	@section S0_1_contact Contact / Support
				16
				17	Please email developer@arm.com
				18
				19	In order to facilitate the work of the support team please provide the build information of the library you are using. To get the version of the library you are using simply run:
				20
				21	$ strings android-armv7a-cl-asserts/libarm_compute.so \| grep arm_compute_version
				22	arm_compute_version=v16.12 Build options: {'embed_kernels': '1', 'opencl': '1', 'arch': 'armv7a', 'neon': '0', 'asserts': '1', 'debug': '0', 'os': 'android', 'Werror': '1'} Git hash=f51a545d4ea12a9059fe4e598a092f1fd06dc858
				23
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	24	@section S0_2_prebuilt_binaries Pre-built binaries
				25
				26	For each release we provide some pre-built binaries of the library [here](https://github.com/ARM-software/ComputeLibrary/releases)
				27
				28	These binaries have been built using the following toolchains:
				29	- Linux armv7a: gcc-linaro-arm-linux-gnueabihf-4.9-2014.07_linux
				30	- Linux arm64-v8a: gcc-linaro-4.9-2016.02-x86_64_aarch64-linux-gnu
Anthony Barbier	a8a28f6	2018-02-26 19:16:32 +0000	[diff] [blame]	31	- Android armv7a: clang++ / gnustl NDK r16b
				32	- Android am64-v8a: clang++ / gnustl NDK r16b
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	33
				34	@warning Make sure to use a compatible toolchain to build your application or you will get some std::bad_alloc errors at runtime.
				35
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	36	@section S1_file_organisation File organisation
				37
				38	This archive contains:
				39	- The arm_compute header and source files
				40	- The latest Khronos OpenCL 1.2 C headers from the <a href="https://www.khronos.org/registry/cl/">Khronos OpenCL registry</a>
				41	- The latest Khronos cl2.hpp from the <a href="https://www.khronos.org/registry/cl/">Khronos OpenCL registry</a> (API version 2.1 when this document was written)
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	42	- The latest Khronos OpenGL ES 3.1 C headers from the <a href="https://www.khronos.org/registry/gles/">Khronos OpenGL ES registry</a>
				43	- The latest Khronos EGL 1.5 C headers from the <a href="https://www.khronos.org/registry/gles/">Khronos EGL registry</a>
				44	- The sources for a stub version of libOpenCL.so, libGLESv1_CM.so, libGLESv2.so and libEGL.so to help you build your application.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	45	- An examples folder containing a few examples to compile and link against the library.
				46	- A @ref utils folder containing headers with some boiler plate code used by the examples.
				47	- This documentation.
				48
				49	You should have the following file organisation:
				50
				51	.
				52	├── arm_compute --> All the arm_compute headers
				53	│ ├── core
				54	│ │ ├── CL
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	55	│ │ │ ├── CLKernelLibrary.h --> Manages all the OpenCL kernels compilation and caching, provides accessors for the OpenCL Context.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	56	│ │ │ ├── CLKernels.h --> Includes all the OpenCL kernels at once
				57	│ │ │ ├── CL specialisation of all the generic objects interfaces (ICLTensor, ICLImage, etc.)
				58	│ │ │ ├── kernels --> Folder containing all the OpenCL kernels
				59	│ │ │ │ └── CL*Kernel.h
				60	│ │ │ └── OpenCL.h --> Wrapper to configure the Khronos OpenCL C++ header
				61	│ │ ├── CPP
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	62	│ │ │ ├── CPPKernels.h --> Includes all the CPP kernels at once
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	63	│ │ │ └── kernels --> Folder containing all the CPP kernels
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	64	│ │ │ └── CPP*Kernel.h
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	65	│ │ ├── GLES_COMPUTE
				66	│ │ │ ├── GCKernelLibrary.h --> Manages all the GLES kernels compilation and caching, provides accessors for the GLES Context.
				67	│ │ │ ├── GCKernels.h --> Includes all the GLES kernels at once
				68	│ │ │ ├── GLES specialisation of all the generic objects interfaces (IGCTensor, IGCImage, etc.)
				69	│ │ │ ├── kernels --> Folder containing all the GLES kernels
				70	│ │ │ │ └── GC*Kernel.h
				71	│ │ │ └── OpenGLES.h --> Wrapper to configure the Khronos EGL and OpenGL ES C header
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	72	│ │ ├── NEON
				73	│ │ │ ├── kernels --> Folder containing all the NEON kernels
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	74	│ │ │ │ ├── arm64 --> Folder containing the interfaces for the assembly arm64 NEON kernels
				75	│ │ │ │ ├── arm32 --> Folder containing the interfaces for the assembly arm32 NEON kernels
				76	│ │ │ │ ├── assembly --> Folder containing the NEON assembly routines.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	77	│ │ │ │ └── NE*Kernel.h
				78	│ │ │ └── NEKernels.h --> Includes all the NEON kernels at once
				79	│ │ ├── All common basic types (Types.h, Window, Coordinates, Iterator, etc.)
				80	│ │ ├── All generic objects interfaces (ITensor, IImage, etc.)
				81	│ │ └── Objects metadata classes (ImageInfo, TensorInfo, MultiImageInfo)
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	82	│ ├── graph
				83	│ │ ├── CL --> OpenCL specific operations
				84	│ │ │ └── CLMap.h / CLUnmap.h
				85	│ │ ├── nodes
				86	│ │ │ └── The various nodes supported by the graph API
				87	│ │ ├── Nodes.h --> Includes all the Graph nodes at once.
				88	│ │ └── Graph objects ( INode, ITensorAccessor, Graph, etc.)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	89	│ └── runtime
				90	│ ├── CL
				91	│ │ ├── CL objects & allocators (CLArray, CLImage, CLTensor, etc.)
				92	│ │ ├── functions --> Folder containing all the OpenCL functions
				93	│ │ │ └── CL*.h
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	94	│ │ ├── CLScheduler.h --> Interface to enqueue OpenCL kernels and get/set the OpenCL CommandQueue and ICLTuner.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	95	│ │ └── CLFunctions.h --> Includes all the OpenCL functions at once
				96	│ ├── CPP
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	97	│ │ ├── CPPKernels.h --> Includes all the CPP functions at once.
				98	│ │ └── CPPScheduler.h --> Basic pool of threads to execute CPP/NEON code on several cores in parallel
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	99	│ ├── GLES_COMPUTE
				100	│ │ ├── GLES objects & allocators (GCArray, GCImage, GCTensor, etc.)
				101	│ │ ├── functions --> Folder containing all the GLES functions
				102	│ │ │ └── GC*.h
				103	│ │ ├── GCScheduler.h --> Interface to enqueue GLES kernels and get/set the GLES CommandQueue.
				104	│ │ └── GCFunctions.h --> Includes all the GLES functions at once
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	105	│ ├── NEON
				106	│ │ ├── functions --> Folder containing all the NEON functions
				107	│ │ │ └── NE*.h
				108	│ │ └── NEFunctions.h --> Includes all the NEON functions at once
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	109	│ ├── OMP
				110	│ │ └── OMPScheduler.h --> OpenMP scheduler (Alternative to the CPPScheduler)
				111	│ ├── Memory manager files (LifetimeManager, PoolManager, etc.)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	112	│ └── Basic implementations of the generic object interfaces (Array, Image, Tensor, etc.)
Anthony Barbier	a8a28f6	2018-02-26 19:16:32 +0000	[diff] [blame]	113	├── data -> Contains test images and reference data dumps used by validation tests
				114	├── docs -> Contains Doxyfile and Doxygen sources used to generate the HTML pages in the documentation folder.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	115	├── documentation
				116	│ ├── index.xhtml
				117	│ └── ...
				118	├── documentation.xhtml -> documentation/index.xhtml
				119	├── examples
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	120	│ ├── cl_*.cpp --> OpenCL examples
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	121	│ ├── gc_*.cpp --> GLES compute shaders examples
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	122	│ ├── graph_*.cpp --> Graph examples
				123	│ ├── neoncl_*.cpp --> NEON / OpenCL interoperability examples
				124	│ └── neon_*.cpp --> NEON examples
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	125	├── include
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	126	│ ├── CL
				127	│ │ └── Khronos OpenCL C headers and C++ wrapper
				128	│ ├── half --> FP16 library available from http://half.sourceforge.net
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	129	│ ├── libnpy --> Library to load / write npy buffers, available from https://github.com/llohse/libnpy
				130	│ └── linux --> Headers only needed for Linux builds
				131	│ └── Khronos EGL and OpenGLES headers
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	132	├── opencl-1.2-stubs
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	133	│ └── opencl_stubs.c --> OpenCL stubs implementation
				134	├── opengles-3.1-stubs
				135	│ ├── EGL.c --> EGL stubs implementation
				136	│ └── GLESv2.c --> GLESv2 stubs implementation
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	137	├── scripts
				138	│ ├── caffe_data_extractor.py --> Basic script to export weights from Caffe to npy files
				139	│ └── tensorflow_data_extractor.py --> Basic script to export weights from Tensor Flow to npy files
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	140	├── src
				141	│ ├── core
				142	│ │ └── ... (Same structure as headers)
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	143	│ │ ├── CL
				144	│ │ │ └── cl_kernels --> All the OpenCL kernels
				145	│ │ └── GLES_COMPUTE
				146	│ │ └── cs_shaders --> All the OpenGL ES Compute Shaders
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	147	│ ├── graph
				148	│ │ └── ... (Same structure as headers)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	149	│ └── runtime
				150	│ └── ... (Same structure as headers)
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	151	├── support
				152	│ └── Various headers to work around toolchains / platform issues.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	153	├── tests
				154	│ ├── All test related files shared between validation and benchmark
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	155	│ ├── CL --> OpenCL accessors
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	156	│ ├── GLES_COMPUTE --> GLES accessors
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	157	│ ├── NEON --> NEON accessors
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	158	│ ├── benchmark --> Sources for benchmarking
				159	│ │ ├── Benchmark specific files
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	160	│ │ ├── CL --> OpenCL benchmarking tests
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	161	│ │ ├── GLES_COMPUTE --> GLES benchmarking tests
Anthony Barbier	a8a28f6	2018-02-26 19:16:32 +0000	[diff] [blame]	162	│ │ ├── fixtures
				163	│ │ │ └── Fixtures to initialise and run the runtime Functions.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	164	│ │ └── NEON --> NEON benchmarking tests
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	165	│ ├── datasets
				166	│ │ └── Datasets for all the validation / benchmark tests, layer configurations for various networks, etc.
				167	│ ├── framework
				168	│ │ └── Boiler plate code for both validation and benchmark test suites (Command line parsers, instruments, output loggers, etc.)
				169	│ ├── networks
				170	│ │ └── Examples of how to instantiate networks.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	171	│ ├── validation --> Sources for validation
				172	│ │ ├── Validation specific files
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	173	│ │ ├── CL --> OpenCL validation tests
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	174	│ │ ├── GLES_COMPUTE --> GLES validation tests
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	175	│ │ ├── CPP --> C++ reference implementations
				176	│ │ ├── fixtures
				177	│ │ │ └── Fixtures to initialise and run the runtime Functions.
				178	│ │ └── NEON --> NEON validation tests
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	179	│ └── dataset --> Datasets defining common sets of input parameters
				180	└── utils --> Boiler plate code used by examples
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	181	└── Various utilities to print types, load / store assets, etc.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	182
				183	@section S2_versions_changelog Release versions and changelog
				184
				185	@subsection S2_1_versions Release versions
				186
				187	All releases are numbered vYY.MM Where YY are the last two digits of the year, and MM the month number.
				188	If there is more than one release in a month then an extra sequential number is appended at the end:
				189
				190	v17.03 (First release of March 2017)
				191	v17.03.1 (Second release of March 2017)
				192	v17.04 (First release of April 2017)
				193
				194	@note We're aiming at releasing one major public release with new features per quarter. All releases in between will only contain bug fixes.
				195
				196	@subsection S2_2_changelog Changelog
				197
Pablo Tello	eb82fd2	2018-02-23 13:43:50 +0000	[diff] [blame]	198	v18.05 Public maintenance release
				199	- Major redesign in the interface for the neon kernels implemented in assembly.
				200	- Removed arm_compute::NEGEMMLowpAArch64A53Kernel / arm_compute::NEGEMMLowpAArch64Kernel / arm_compute::NEGEMMLowpAArch64V8P4Kernel / arm_compute::NEGEMMInterleavedBlockedKernel / arm_compute::NEGEMMLowpAssemblyMatrixMultiplyCore / arm_compute::NEHGEMMAArch64FP16Kernel
				201	- Added NEGEMMAssemblyWrapper and AssemblyKernelGlue which are used to execute assembly kernels in neon functions.
				202	- Minor changes to the CPUInfo type to make it compatible with the new assembly gemm interface.
				203
Anthony Barbier	577fbdf	2018-03-01 15:17:54 +0000	[diff] [blame]	204	v18.03 Public maintenance release
				205	- Various bug fixes.
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	206	- Fixed bug in @ref NEActivationLayer
				207	- Fix in @ref CLTuner when using batches.
Anthony Barbier	577fbdf	2018-03-01 15:17:54 +0000	[diff] [blame]	208	- Updated recommended NDK version to r16b (And fixed warnings).
				209	- Fixed bug in validation code.
				210	- Added Inception v4 graph example.
				211
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	212	v18.02 Public major release
				213	- Various NEON / OpenCL / GLES optimisations.
				214	- Various bug fixes.
				215	- Changed default number of threads on big LITTLE systems.
				216	- Refactored examples and added:
				217	- graph_mobilenet_qassym8
				218	- graph_resnet
				219	- graph_squeezenet_v1_1
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	220	- Renamed @ref CLConvolutionLayer into @ref CLGEMMConvolutionLayer and created a new @ref CLConvolutionLayer to select the fastest convolution method.
				221	- Renamed @ref NEConvolutionLayer into @ref NEGEMMConvolutionLayer and created a new @ref NEConvolutionLayer to select the fastest convolution method.
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	222	- Added in place support to:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	223	- @ref CLActivationLayer
				224	- @ref CLBatchNormalizationLayer
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	225	- Added QASYMM8 support to:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	226	- @ref CLActivationLayer
				227	- @ref CLDepthwiseConvolutionLayer
				228	- @ref NEDepthwiseConvolutionLayer
				229	- @ref NESoftmaxLayer
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	230	- Added FP16 support to:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	231	- @ref CLDepthwiseConvolutionLayer3x3
				232	- @ref CLDepthwiseConvolutionLayer
				233	- Added broadcasting support to @ref NEArithmeticAddition / @ref CLArithmeticAddition / @ref CLPixelWiseMultiplication
				234	- Added fused batched normalization and activation to @ref CLBatchNormalizationLayer and @ref NEBatchNormalizationLayer
				235	- Added support for non-square pooling to @ref NEPoolingLayer and @ref CLPoolingLayer
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	236	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	237	- @ref CLDirectConvolutionLayerOutputStageKernel
Pablo Tello	f6c572c	2018-02-14 12:47:30 +0000	[diff] [blame]	238	- New NEON kernels / functions
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	239	- Added name() method to all kernels.
				240	- Added support for Winograd 5x5.
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	241	- @ref NEPermuteKernel / @ref NEPermute
				242	- @ref NEWinogradLayerTransformInputKernel / @ref NEWinogradLayer
				243	- @ref NEWinogradLayerTransformOutputKernel / @ref NEWinogradLayer
				244	- @ref NEWinogradLayerTransformWeightsKernel / @ref NEWinogradLayer
				245	- Renamed NEWinogradLayerKernel into @ref NEWinogradLayerBatchedGEMMKernel
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	246	- New GLES kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	247	- @ref GCTensorShiftKernel / @ref GCTensorShift
Pablo Tello	f6c572c	2018-02-14 12:47:30 +0000	[diff] [blame]	248
Anthony Barbier	64c95a0	2018-01-22 18:48:55 +0000	[diff] [blame]	249	v18.01 Public maintenance release
				250	- Various bug fixes
				251	- Added some of the missing validate() methods
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	252	- Added @ref CLDeconvolutionLayerUpsampleKernel / @ref CLDeconvolutionLayer @ref CLDeconvolutionLayerUpsample
				253	- Added @ref CLPermuteKernel / @ref CLPermute
Anthony Barbier	64c95a0	2018-01-22 18:48:55 +0000	[diff] [blame]	254	- Added method to clean the programs cache in the CL Kernel library.
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	255	- Added @ref GCArithmeticAdditionKernel / @ref GCArithmeticAddition
				256	- Added @ref GCDepthwiseConvolutionLayer3x3Kernel / @ref GCDepthwiseConvolutionLayer3x3
				257	- Added @ref GCNormalizePlanarYUVLayerKernel / @ref GCNormalizePlanarYUVLayer
				258	- Added @ref GCScaleKernel / @ref GCScale
				259	- Added @ref GCWeightsReshapeKernel / @ref GCConvolutionLayer
Anthony Barbier	64c95a0	2018-01-22 18:48:55 +0000	[diff] [blame]	260	- Added FP16 support to the following GLES compute kernels:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	261	- @ref GCCol2ImKernel
				262	- @ref GCGEMMInterleave4x4Kernel
				263	- @ref GCGEMMTranspose1xWKernel
				264	- @ref GCIm2ColKernel
				265	- Refactored NEON Winograd (NEWinogradLayerKernel)
				266	- Added @ref NEDirectConvolutionLayerOutputStageKernel
Anthony Barbier	64c95a0	2018-01-22 18:48:55 +0000	[diff] [blame]	267	- Added QASYMM8 support to the following NEON kernels:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	268	- @ref NEDepthwiseConvolutionLayer3x3Kernel
				269	- @ref NEFillBorderKernel
				270	- @ref NEPoolingLayerKernel
Anthony Barbier	64c95a0	2018-01-22 18:48:55 +0000	[diff] [blame]	271	- Added new examples:
				272	- graph_cl_mobilenet_qasymm8.cpp
				273	- graph_inception_v3.cpp
				274	- gc_dc.cpp
				275	- More tests added to both validation and benchmarking suites.
				276
Gian Marco	ff85093	2017-12-11 12:37:17 +0000	[diff] [blame]	277	v17.12 Public major release
				278	- Most machine learning functions on OpenCL support the new data type QASYMM8
				279	- Introduced logging interface
				280	- Introduced opencl timer
				281	- Reworked GEMMLowp interface
				282	- Added new NEON assembly kernels for GEMMLowp, SGEMM and HGEMM
				283	- Added validation method for most Machine Learning kernels / functions
				284	- Added new graph examples such as googlenet, mobilenet, squeezenet, vgg16 and vgg19
				285	- Added sgemm example for OpenCL
				286	- Added absolute difference example for GLES compute
				287	- Added new tests and benchmarks in validation and benchmark frameworks
				288	- Added new kernels / functions for GLES compute
				289
				290	- New OpenGL ES kernels / functions
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	291	- @ref GCAbsoluteDifferenceKernel / @ref GCAbsoluteDifference
				292	- @ref GCActivationLayerKernel / @ref GCActivationLayer
				293	- @ref GCBatchNormalizationLayerKernel / @ref GCBatchNormalizationLayer
				294	- @ref GCCol2ImKernel
				295	- @ref GCDepthConcatenateLayerKernel / @ref GCDepthConcatenateLayer
				296	- @ref GCDirectConvolutionLayerKernel / @ref GCDirectConvolutionLayer
				297	- @ref GCDropoutLayerKernel / @ref GCDropoutLayer
				298	- @ref GCFillBorderKernel / @ref GCFillBorder
				299	- @ref GCGEMMInterleave4x4Kernel / @ref GCGEMMInterleave4x4
				300	- @ref GCGEMMMatrixAccumulateBiasesKernel / @ref GCGEMMMatrixAdditionKernel / @ref GCGEMMMatrixMultiplyKernel / @ref GCGEMM
				301	- @ref GCGEMMTranspose1xWKernel / @ref GCGEMMTranspose1xW
				302	- @ref GCIm2ColKernel
				303	- @ref GCNormalizationLayerKernel / @ref GCNormalizationLayer
				304	- @ref GCPixelWiseMultiplicationKernel / @ref GCPixelWiseMultiplication
				305	- @ref GCPoolingLayerKernel / @ref GCPoolingLayer
				306	- @ref GCLogits1DMaxKernel / @ref GCLogits1DShiftExpSumKernel / @ref GCLogits1DNormKernel / @ref GCSoftmaxLayer
				307	- @ref GCTransposeKernel / @ref GCTranspose
Gian Marco	ff85093	2017-12-11 12:37:17 +0000	[diff] [blame]	308
				309	- New NEON kernels / functions
Pablo Tello	eb82fd2	2018-02-23 13:43:50 +0000	[diff] [blame]	310	- arm_compute::NEGEMMLowpAArch64A53Kernel / arm_compute::NEGEMMLowpAArch64Kernel / arm_compute::NEGEMMLowpAArch64V8P4Kernel / arm_compute::NEGEMMInterleavedBlockedKernel / arm_compute::NEGEMMLowpAssemblyMatrixMultiplyCore
				311	- arm_compute::NEHGEMMAArch64FP16Kernel
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	312	- @ref NEDepthwiseConvolutionLayer3x3Kernel / @ref NEDepthwiseIm2ColKernel / @ref NEGEMMMatrixVectorMultiplyKernel / @ref NEDepthwiseVectorToTensorKernel / @ref NEDepthwiseConvolutionLayer
				313	- @ref NEGEMMLowpOffsetContributionKernel / @ref NEGEMMLowpMatrixAReductionKernel / @ref NEGEMMLowpMatrixBReductionKernel / @ref NEGEMMLowpMatrixMultiplyCore
				314	- @ref NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel / @ref NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint
				315	- @ref NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel / @ref NEGEMMLowpQuantizeDownInt32ToUint8Scale
				316	- @ref NEWinogradLayer / NEWinogradLayerKernel
Gian Marco	ff85093	2017-12-11 12:37:17 +0000	[diff] [blame]	317
				318	- New OpenCL kernels / functions
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	319	- @ref CLGEMMLowpOffsetContributionKernel / @ref CLGEMMLowpMatrixAReductionKernel / @ref CLGEMMLowpMatrixBReductionKernel / @ref CLGEMMLowpMatrixMultiplyCore
				320	- @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel / @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint
				321	- @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleKernel / @ref CLGEMMLowpQuantizeDownInt32ToUint8Scale
Gian Marco	ff85093	2017-12-11 12:37:17 +0000	[diff] [blame]	322
				323	- New graph nodes for NEON and OpenCL
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	324	- graph::BranchLayer
				325	- graph::DepthConvertLayer
				326	- graph::DepthwiseConvolutionLayer
				327	- graph::DequantizationLayer
				328	- graph::FlattenLayer
				329	- graph::QuantizationLayer
				330	- graph::ReshapeLayer
Gian Marco	ff85093	2017-12-11 12:37:17 +0000	[diff] [blame]	331
Anthony Barbier	3c5b4ff	2017-10-12 13:20:52 +0100	[diff] [blame]	332	v17.10 Public maintenance release
				333	- Bug fixes:
				334	- Check the maximum local workgroup size supported by OpenCL devices
				335	- Minor documentation updates (Fixed instructions to build the examples)
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	336	- Introduced a graph::GraphContext
Anthony Barbier	3c5b4ff	2017-10-12 13:20:52 +0100	[diff] [blame]	337	- Added a few new Graph nodes, support for branches and grouping.
				338	- Automatically enable cl_printf in debug builds
				339	- Fixed bare metal builds for armv7a
				340	- Added AlexNet and cartoon effect examples
				341	- Fixed library builds: libraries are no longer built as supersets of each other.(It means application using the Runtime part of the library now need to link against both libarm_compute_core and libarm_compute)
				342
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	343	v17.09 Public major release
				344	- Experimental Graph support: initial implementation of a simple stream API to easily chain machine learning layers.
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	345	- Memory Manager (@ref BlobLifetimeManager, @ref BlobMemoryPool, @ref ILifetimeManager, @ref IMemoryGroup, @ref IMemoryManager, @ref IMemoryPool, @ref IPoolManager, @ref MemoryManagerOnDemand, @ref PoolManager)
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	346	- New validation and benchmark frameworks (Boost and Google frameworks replaced by homemade framework).
				347	- Most machine learning functions support both fixed point 8 and 16 bit (QS8, QS16) for both NEON and OpenCL.
				348	- New NEON kernels / functions:
Pablo Tello	eb82fd2	2018-02-23 13:43:50 +0000	[diff] [blame]	349	- arm_compute::NEGEMMAssemblyBaseKernel arm_compute::NEGEMMAArch64Kernel
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	350	- @ref NEDequantizationLayerKernel / @ref NEDequantizationLayer
				351	- @ref NEFloorKernel / @ref NEFloor
				352	- @ref NEL2NormalizeLayerKernel / @ref NEL2NormalizeLayer
				353	- @ref NEQuantizationLayerKernel @ref NEMinMaxLayerKernel / @ref NEQuantizationLayer
				354	- @ref NEROIPoolingLayerKernel / @ref NEROIPoolingLayer
				355	- @ref NEReductionOperationKernel / @ref NEReductionOperation
				356	- @ref NEReshapeLayerKernel / @ref NEReshapeLayer
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	357
				358	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	359	- @ref CLDepthwiseConvolutionLayer3x3Kernel @ref CLDepthwiseIm2ColKernel @ref CLDepthwiseVectorToTensorKernel @ref CLDepthwiseWeightsReshapeKernel / @ref CLDepthwiseConvolutionLayer3x3 @ref CLDepthwiseConvolutionLayer @ref CLDepthwiseSeparableConvolutionLayer
				360	- @ref CLDequantizationLayerKernel / @ref CLDequantizationLayer
				361	- @ref CLDirectConvolutionLayerKernel / @ref CLDirectConvolutionLayer
				362	- @ref CLFlattenLayer
				363	- @ref CLFloorKernel / @ref CLFloor
				364	- @ref CLGEMMTranspose1xW
				365	- @ref CLGEMMMatrixVectorMultiplyKernel
				366	- @ref CLL2NormalizeLayerKernel / @ref CLL2NormalizeLayer
				367	- @ref CLQuantizationLayerKernel @ref CLMinMaxLayerKernel / @ref CLQuantizationLayer
				368	- @ref CLROIPoolingLayerKernel / @ref CLROIPoolingLayer
				369	- @ref CLReductionOperationKernel / @ref CLReductionOperation
				370	- @ref CLReshapeLayerKernel / @ref CLReshapeLayer
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	371
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	372	v17.06 Public major release
				373	- Various bug fixes
				374	- Added support for fixed point 8 bit (QS8) to the various NEON machine learning kernels.
				375	- Added unit tests and benchmarks (AlexNet, LeNet)
				376	- Added support for sub tensors.
				377	- Added infrastructure to provide GPU specific optimisation for some OpenCL kernels.
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	378	- Added @ref OMPScheduler (OpenMP) scheduler for NEON
				379	- Added @ref SingleThreadScheduler scheduler for NEON (For bare metal)
				380	- User can specify his own scheduler by implementing the @ref IScheduler interface.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	381	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	382	- @ref CLBatchNormalizationLayerKernel / @ref CLBatchNormalizationLayer
				383	- @ref CLDepthConcatenateLayerKernel / @ref CLDepthConcatenateLayer
				384	- @ref CLHOGOrientationBinningKernel @ref CLHOGBlockNormalizationKernel, @ref CLHOGDetectorKernel / @ref CLHOGDescriptor @ref CLHOGDetector @ref CLHOGGradient @ref CLHOGMultiDetection
				385	- @ref CLLocallyConnectedMatrixMultiplyKernel / @ref CLLocallyConnectedLayer
				386	- @ref CLWeightsReshapeKernel / @ref CLConvolutionLayerReshapeWeights
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	387	- New C++ kernels:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	388	- @ref CPPDetectionWindowNonMaximaSuppressionKernel
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	389	- New NEON kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	390	- @ref NEBatchNormalizationLayerKernel / @ref NEBatchNormalizationLayer
				391	- @ref NEDepthConcatenateLayerKernel / @ref NEDepthConcatenateLayer
				392	- @ref NEDirectConvolutionLayerKernel / @ref NEDirectConvolutionLayer
				393	- @ref NELocallyConnectedMatrixMultiplyKernel / @ref NELocallyConnectedLayer
				394	- @ref NEWeightsReshapeKernel / @ref NEConvolutionLayerReshapeWeights
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	395
				396	v17.05 Public bug fixes release
				397	- Various bug fixes
				398	- Remaining of the functions ported to use accurate padding.
				399	- Library does not link against OpenCL anymore (It uses dlopen / dlsym at runtime instead to determine whether or not OpenCL is available).
				400	- Added "free" method to allocator.
				401	- Minimum version of g++ required for armv7 Linux changed from 4.8 to 4.9
				402
				403	v17.04 Public bug fixes release
				404
				405	The following functions have been ported to use the new accurate padding:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	406	- @ref CLColorConvertKernel
				407	- @ref CLEdgeNonMaxSuppressionKernel
				408	- @ref CLEdgeTraceKernel
				409	- @ref CLGaussianPyramidHorKernel
				410	- @ref CLGaussianPyramidVertKernel
				411	- @ref CLGradientKernel
				412	- @ref NEChannelCombineKernel
				413	- @ref NEFillArrayKernel
				414	- @ref NEGaussianPyramidHorKernel
				415	- @ref NEGaussianPyramidVertKernel
				416	- @ref NEHarrisScoreFP16Kernel
				417	- @ref NEHarrisScoreKernel
				418	- @ref NEHOGDetectorKernel
				419	- @ref NELogits1DMaxKernel
				420	- NELogits1DShiftExpSumKernel
				421	- NELogits1DNormKernel
				422	- @ref NENonMaximaSuppression3x3FP16Kernel
				423	- @ref NENonMaximaSuppression3x3Kernel
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	424
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	425	v17.03.1 First Major public release of the sources
				426	- Renamed the library to arm_compute
				427	- New CPP target introduced for C++ kernels shared between NEON and CL functions.
				428	- New padding calculation interface introduced and ported most kernels / functions to use it.
				429	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	430	- @ref CLGEMMLowpMatrixMultiplyKernel / CLGEMMLowp
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	431	- New NEON kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	432	- @ref NENormalizationLayerKernel / @ref NENormalizationLayer
				433	- @ref NETransposeKernel / @ref NETranspose
				434	- @ref NELogits1DMaxKernel, NELogits1DShiftExpSumKernel, NELogits1DNormKernel / @ref NESoftmaxLayer
				435	- @ref NEIm2ColKernel, @ref NECol2ImKernel, NEConvolutionLayerWeightsReshapeKernel / @ref NEConvolutionLayer
				436	- @ref NEGEMMMatrixAccumulateBiasesKernel / @ref NEFullyConnectedLayer
				437	- @ref NEGEMMLowpMatrixMultiplyKernel / NEGEMMLowp
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	438
				439	v17.03 Sources preview
				440	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	441	- @ref CLGradientKernel, @ref CLEdgeNonMaxSuppressionKernel, @ref CLEdgeTraceKernel / @ref CLCannyEdge
				442	- GEMM refactoring + FP16 support: @ref CLGEMMInterleave4x4Kernel, @ref CLGEMMTranspose1xWKernel, @ref CLGEMMMatrixMultiplyKernel, @ref CLGEMMMatrixAdditionKernel / @ref CLGEMM
				443	- @ref CLGEMMMatrixAccumulateBiasesKernel / @ref CLFullyConnectedLayer
				444	- @ref CLTransposeKernel / @ref CLTranspose
				445	- @ref CLLKTrackerInitKernel, @ref CLLKTrackerStage0Kernel, @ref CLLKTrackerStage1Kernel, @ref CLLKTrackerFinalizeKernel / @ref CLOpticalFlow
				446	- @ref CLNormalizationLayerKernel / @ref CLNormalizationLayer
				447	- @ref CLLaplacianPyramid, @ref CLLaplacianReconstruct
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	448	- New NEON kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	449	- @ref NEActivationLayerKernel / @ref NEActivationLayer
				450	- GEMM refactoring + FP16 support (Requires armv8.2 CPU): @ref NEGEMMInterleave4x4Kernel, @ref NEGEMMTranspose1xWKernel, @ref NEGEMMMatrixMultiplyKernel, @ref NEGEMMMatrixAdditionKernel / @ref NEGEMM
				451	- @ref NEPoolingLayerKernel / @ref NEPoolingLayer
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	452
				453	v17.02.1 Sources preview
				454	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	455	- @ref CLLogits1DMaxKernel, @ref CLLogits1DShiftExpSumKernel, @ref CLLogits1DNormKernel / @ref CLSoftmaxLayer
				456	- @ref CLPoolingLayerKernel / @ref CLPoolingLayer
				457	- @ref CLIm2ColKernel, @ref CLCol2ImKernel, CLConvolutionLayerWeightsReshapeKernel / @ref CLConvolutionLayer
				458	- @ref CLRemapKernel / @ref CLRemap
				459	- @ref CLGaussianPyramidHorKernel, @ref CLGaussianPyramidVertKernel / @ref CLGaussianPyramid, @ref CLGaussianPyramidHalf, @ref CLGaussianPyramidOrb
				460	- @ref CLMinMaxKernel, @ref CLMinMaxLocationKernel / @ref CLMinMaxLocation
				461	- @ref CLNonLinearFilterKernel / @ref CLNonLinearFilter
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	462	- New NEON FP16 kernels (Requires armv8.2 CPU)
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	463	- @ref NEAccumulateWeightedFP16Kernel
				464	- @ref NEBox3x3FP16Kernel
				465	- @ref NENonMaximaSuppression3x3FP16Kernel
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	466
				467	v17.02 Sources preview
				468	- New OpenCL kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	469	- @ref CLActivationLayerKernel / @ref CLActivationLayer
				470	- @ref CLChannelCombineKernel / @ref CLChannelCombine
				471	- @ref CLDerivativeKernel / @ref CLChannelExtract
				472	- @ref CLFastCornersKernel / @ref CLFastCorners
				473	- @ref CLMeanStdDevKernel / @ref CLMeanStdDev
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	474	- New NEON kernels / functions:
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	475	- HOG / SVM: @ref NEHOGOrientationBinningKernel, @ref NEHOGBlockNormalizationKernel, @ref NEHOGDetectorKernel, NEHOGNonMaximaSuppressionKernel / @ref NEHOGDescriptor, @ref NEHOGDetector, @ref NEHOGGradient, @ref NEHOGMultiDetection
				476	- @ref NENonLinearFilterKernel / @ref NENonLinearFilter
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	477	- Introduced a CLScheduler to manage the default context and command queue used by the runtime library and create synchronisation events.
				478	- Switched all the kernels / functions to use tensors instead of images.
				479	- Updated documentation to include instructions to build the library from sources.
				480
				481	v16.12 Binary preview release
				482	- Original release
				483
				484	@section S3_how_to_build How to build the library and the examples
				485
				486	@subsection S3_1_build_options Build options
				487
				488	scons 2.3 or above is required to build the library.
				489	To see the build options available simply run ```scons -h```:
				490
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	491	debug: Debug (yes\|no)
				492	default: False
				493	actual: False
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	494
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	495	asserts: Enable asserts (this flag is forced to 1 for debug=1) (yes\|no)
				496	default: False
				497	actual: False
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	498
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	499	arch: Target Architecture (armv7a\|arm64-v8a\|arm64-v8.2-a\|x86_32\|x86_64)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	500	default: armv7a
				501	actual: armv7a
				502
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	503	os: Target OS (linux\|android\|bare_metal)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	504	default: linux
				505	actual: linux
				506
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	507	build: Build type (native\|cross_compile\|embed_only)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	508	default: cross_compile
				509	actual: cross_compile
				510
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	511	examples: Build example programs (yes\|no)
				512	default: True
				513	actual: True
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	514
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	515	Werror: Enable/disable the -Werror compilation flag (yes\|no)
				516	default: True
				517	actual: True
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	518
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	519	opencl: Enable OpenCL support (yes\|no)
				520	default: True
				521	actual: True
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	522
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	523	neon: Enable Neon support (yes\|no)
				524	default: False
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	525	actual: False
				526
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	527	gles_compute: Enable OpenGL ES Compute Shader support (yes\|no)
				528	default: False
				529	actual: False
				530
				531	embed_kernels: Embed OpenCL kernels and OpenGL ES compute shader in library binary (yes\|no)
Anthony Barbier	cc0a80b	2017-12-15 11:37:29 +0000	[diff] [blame]	532	default: True
				533	actual: True
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	534
				535	set_soname: Set the library's soname and shlibversion (requires SCons 2.4 or above) (yes\|no)
				536	default: False
				537	actual: False
				538
				539	openmp: Enable OpenMP backend (yes\|no)
				540	default: False
				541	actual: False
				542
				543	cppthreads: Enable C++11 threads backend (yes\|no)
				544	default: True
				545	actual: True
				546
				547	build_dir: Specify sub-folder for the build ( /path/to/build_dir )
				548	default: .
				549	actual: .
				550
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	551	extra_cxx_flags: Extra CXX flags to be appended to the build command
				552	default:
				553	actual:
				554
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	555	pmu: Enable PMU counters (yes\|no)
				556	default: False
				557	actual: False
				558
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	559	mali: Enable Mali hardware counters (yes\|no)
				560	default: False
				561	actual: False
				562
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	563	validation_tests: Build validation test programs (yes\|no)
				564	default: False
				565	actual: False
				566
				567	benchmark_tests: Build benchmark test programs (yes\|no)
				568	default: False
				569	actual: False
				570
				571	@b debug / @b asserts:
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	572	- With debug=1 asserts are enabled, and the library is built with symbols and no optimisations enabled.
				573	- With debug=0 and asserts=1: Optimisations are enabled and symbols are removed, however all the asserts are still present (This is about 20% slower than the release build)
				574	- With debug=0 and asserts=0: All optimisations are enable and no validation is performed, if the application misuses the library it is likely to result in a crash. (Only use this mode once you are sure your application is working as expected).
				575
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	576	@b arch: The x86_32 and x86_64 targets can only be used with neon=0 and opencl=1.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	577
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	578	@b os: Choose the operating system you are targeting: Linux, Android or bare metal.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	579	@note bare metal can only be used for NEON (not OpenCL), only static libraries get built and NEON's multi-threading support is disabled.
				580
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	581	@b build: you can either build directly on your device (native) or cross compile from your desktop machine (cross-compile). In both cases make sure the compiler is available in your path.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	582
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	583	@note If you want to natively compile for 32bit on a 64bit ARM device running a 64bit OS then you will have to use cross-compile too.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	584
Anthony Barbier	2d0ce77	2018-02-21 15:35:36 +0000	[diff] [blame]	585	There is also an 'embed_only' option which will generate all the .embed files for the OpenCL kernels and / or OpenGLES compute shaders. This might be useful if using a different build system to compile the library.
				586
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	587	@b Werror: If you are compiling using the same toolchains as the ones used in this guide then there shouldn't be any warning and therefore you should be able to keep Werror=1. If with a different compiler version the library fails to build because of warnings interpreted as errors then, if you are sure the warnings are not important, you might want to try to build with Werror=0 (But please do report the issue either on Github or by an email to developer@arm.com so that the issue can be addressed).
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	588
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	589	@b opencl / @b neon / @b gles_compute: Choose which SIMD technology you want to target. (NEON for ARM Cortex-A CPUs or OpenCL / GLES_COMPUTE for ARM Mali GPUs)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	590
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	591	@b embed_kernels: For OpenCL / GLES_COMPUTE only: set embed_kernels=1 if you want the OpenCL / GLES_COMPUTE kernels to be built in the library's binaries instead of being read from separate ".cl" / ".cs" files. If embed_kernels is set to 0 then the application can set the path to the folder containing the OpenCL / GLES_COMPUTE kernel files by calling CLKernelLibrary::init() / GCKernelLibrary::init(). By default the path is set to "./cl_kernels" / "./cs_shaders".
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	592
				593	@b set_soname: Do you want to build the versioned version of the library ?
				594
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	595	If enabled the library will contain a SONAME and SHLIBVERSION and some symlinks will automatically be created between the objects.
				596	Example:
				597	libarm_compute_core.so -> libarm_compute_core.so.1.0.0
				598	libarm_compute_core.so.1 -> libarm_compute_core.so.1.0.0
				599	libarm_compute_core.so.1.0.0
				600
				601	@note This options is disabled by default as it requires SCons version 2.4 or above.
				602
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	603	@b extra_cxx_flags: Custom CXX flags which will be appended to the end of the build command.
				604
				605	@b build_dir: Build the library in a subfolder of the "build" folder. (Allows to build several configurations in parallel).
				606
				607	@b examples: Build or not the examples
				608
				609	@b validation_tests: Enable the build of the validation suite.
				610
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	611	@b benchmark_tests: Enable the build of the benchmark tests
				612
				613	@b pmu: Enable the PMU cycle counter to measure execution time in benchmark tests. (Your device needs to support it)
				614
Anthony Barbier	6a5627a	2017-09-26 14:42:02 +0100	[diff] [blame]	615	@b mali: Enable the collection of Mali hardware counters to measure execution time in benchmark tests. (Your device needs to have a Mali driver that supports it)
				616
Anthony Barbier	79c6178	2017-06-23 11:48:24 +0100	[diff] [blame]	617	@b openmp Build in the OpenMP scheduler for NEON.
				618
				619	@note Only works when building with g++ not clang++
				620
				621	@b cppthreads Build in the C++11 scheduler for NEON.
				622
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	623	@sa Scheduler::set
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	624
Moritz Pflanzer	07674de	2017-07-21 09:39:36 +0100	[diff] [blame]	625	@subsection S3_2_linux Building for Linux
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	626
				627	@subsubsection S3_2_1_library How to build the library ?
				628
				629	For Linux, the library was successfully built and tested using the following Linaro GCC toolchain:
				630
				631	- gcc-linaro-arm-linux-gnueabihf-4.9-2014.07_linux
				632	- gcc-linaro-4.9-2016.02-x86_64_aarch64-linux-gnu
				633	- gcc-linaro-6.3.1-2017.02-i686_aarch64-linux-gnu
				634
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	635	To cross-compile the library in debug mode, with NEON only support, for Linux 32bit:
				636
				637	scons Werror=1 -j8 debug=1 neon=1 opencl=0 os=linux arch=armv7a
				638
				639	To cross-compile the library in asserts mode, with OpenCL only support, for Linux 64bit:
				640
				641	scons Werror=1 -j8 debug=0 asserts=1 neon=0 opencl=1 embed_kernels=1 os=linux arch=arm64-v8a
				642
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	643	To cross-compile the library in asserts mode, with GLES_COMPUTE only support, for Linux 64bit:
				644
				645	scons Werror=1 -j8 debug=0 asserts=1 neon=0 opencl=0 gles_compute=1 embed_kernels=1 os=linux arch=arm64-v8a
				646
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	647	You can also compile the library natively on an ARM device by using <b>build=native</b>:
				648
				649	scons Werror=1 -j8 debug=0 neon=1 opencl=0 os=linux arch=arm64-v8a build=native
				650	scons Werror=1 -j8 debug=0 neon=1 opencl=0 os=linux arch=armv7a build=native
				651
				652	@note g++ for ARM is mono-arch, therefore if you want to compile for Linux 32bit on a Linux 64bit platform you will have to use a cross compiler.
				653
				654	For example on a 64bit Debian based system you would have to install <b>g++-arm-linux-gnueabihf</b>
				655
				656	apt-get install g++-arm-linux-gnueabihf
				657
				658	Then run
				659
				660	scons Werror=1 -j8 debug=0 neon=1 opencl=0 os=linux arch=armv7a build=cross_compile
				661
				662	or simply remove the build parameter as build=cross_compile is the default value:
				663
				664	scons Werror=1 -j8 debug=0 neon=1 opencl=0 os=linux arch=armv7a
				665
				666	@attention To cross compile with opencl=1 you need to make sure to have a version of libOpenCL matching your target architecture.
				667
				668	@subsubsection S3_2_2_examples How to manually build the examples ?
				669
				670	The examples get automatically built by scons as part of the build process of the library described above. This section just describes how you can build and link your own application against our library.
				671
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	672	@note The following command lines assume the arm_compute binaries are present in the current directory or in the system library path. If this is not the case you can specify the location of the pre-built library with the compiler option -L. When building the OpenCL example the commands below assume that the CL headers are located in the include folder where the command is executed.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	673
				674	To cross compile a NEON example for Linux 32bit:
				675
Anthony Barbier	b2881fc	2017-09-29 17:12:12 +0100	[diff] [blame]	676	arm-linux-gnueabihf-g++ examples/neon_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -mfpu=neon -L. -larm_compute -larm_compute_core -o neon_convolution
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	677
				678	To cross compile a NEON example for Linux 64bit:
				679
Anthony Barbier	b2881fc	2017-09-29 17:12:12 +0100	[diff] [blame]	680	aarch64-linux-gnu-g++ examples/neon_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -L. -larm_compute -larm_compute_core -o neon_convolution
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	681
				682	(notice the only difference with the 32 bit command is that we don't need the -mfpu option and the compiler's name is different)
				683
				684	To cross compile an OpenCL example for Linux 32bit:
				685
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	686	arm-linux-gnueabihf-g++ examples/cl_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -mfpu=neon -L. -larm_compute -larm_compute_core -o cl_convolution -DARM_COMPUTE_CL
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	687
				688	To cross compile an OpenCL example for Linux 64bit:
				689
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	690	aarch64-linux-gnu-g++ examples/cl_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -L. -larm_compute -larm_compute_core -o cl_convolution -DARM_COMPUTE_CL
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	691
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	692	To cross compile a GLES example for Linux 32bit:
				693
				694	arm-linux-gnueabihf-g++ examples/gc_absdiff.cpp utils/Utils.cpp -I. -Iinclude/ -L. -larm_compute -larm_compute_core -std=c++11 -mfpu=neon -DARM_COMPUTE_GC -Iinclude/linux/ -o gc_absdiff
				695
				696	To cross compile a GLES example for Linux 64bit:
				697
				698	aarch64-linux-gnu-g++ examples/gc_absdiff.cpp utils/Utils.cpp -I. -Iinclude/ -L. -larm_compute -larm_compute_core -std=c++11 -DARM_COMPUTE_GC -Iinclude/linux/ -o gc_absdiff
				699
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	700	(notice the only difference with the 32 bit command is that we don't need the -mfpu option and the compiler's name is different)
				701
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	702	To cross compile the examples with the Graph API, such as graph_lenet.cpp, you need to link the examples against arm_compute_graph.so too.
				703
				704	@note The compute library must currently be built with both neon and opencl enabled - neon=1 and opencl=1
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	705
				706	i.e. to cross compile the "graph_lenet" example for Linux 32bit:
				707
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	708	arm-linux-gnueabihf-g++ examples/graph_lenet.cpp utils/Utils.cpp utils/GraphUtils.cpp -I. -Iinclude -std=c++11 -mfpu=neon -L. -larm_compute_graph -larm_compute -larm_compute_core -Wl,--allow-shlib-undefined -o graph_lenet
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	709
				710	i.e. to cross compile the "graph_lenet" example for Linux 64bit:
				711
Isabella Gottardi	b28f29d	2017-11-09 17:05:07 +0000	[diff] [blame]	712	aarch64-linux-gnu-g++ examples/graph_lenet.cpp utils/Utils.cpp utils/GraphUtils.cpp -I. -Iinclude -std=c++11 -L. -larm_compute_graph -larm_compute -larm_compute_core -Wl,--allow-shlib-undefined -o graph_lenet
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	713
				714	(notice the only difference with the 32 bit command is that we don't need the -mfpu option and the compiler's name is different)
				715
Anthony Barbier	e500747	2017-10-27 15:01:44 +0100	[diff] [blame]	716	@note If compiling using static libraries, this order must be followed when linking: arm_compute_graph_static, arm_compute, arm_compute_core
				717
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	718	To compile natively (i.e directly on an ARM device) for NEON for Linux 32bit:
				719
Anthony Barbier	b2881fc	2017-09-29 17:12:12 +0100	[diff] [blame]	720	g++ examples/neon_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -mfpu=neon -larm_compute -larm_compute_core -o neon_convolution
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	721
				722	To compile natively (i.e directly on an ARM device) for NEON for Linux 64bit:
				723
Anthony Barbier	b2881fc	2017-09-29 17:12:12 +0100	[diff] [blame]	724	g++ examples/neon_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute -larm_compute_core -o neon_convolution
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	725
				726	(notice the only difference with the 32 bit command is that we don't need the -mfpu option)
				727
				728	To compile natively (i.e directly on an ARM device) for OpenCL for Linux 32bit or Linux 64bit:
				729
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	730	g++ examples/cl_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute -larm_compute_core -o cl_convolution -DARM_COMPUTE_CL
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	731
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	732	To compile natively (i.e directly on an ARM device) for GLES for Linux 32bit or Linux 64bit:
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	733
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	734	g++ examples/gc_absdiff.cpp utils/Utils.cpp -I. -Iinclude/ -L. -larm_compute -larm_compute_core -std=c++11 -DARM_COMPUTE_GC -Iinclude/linux/ -o gc_absdiff
				735
				736	To compile natively the examples with the Graph API, such as graph_lenet.cpp, you need to link the examples against arm_compute_graph.so too.
				737	@note The compute library must currently be built with both neon and opencl enabled - neon=1 and opencl=1
				738
				739	i.e. to natively compile the "graph_lenet" example for Linux 32bit:
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	740
Isabella Gottardi	b28f29d	2017-11-09 17:05:07 +0000	[diff] [blame]	741	g++ examples/graph_lenet.cpp utils/Utils.cpp utils/GraphUtils.cpp -I. -Iinclude -std=c++11 -mfpu=neon -L. -larm_compute_graph -larm_compute -larm_compute_core -Wl,--allow-shlib-undefined -o graph_lenet
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	742
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	743	i.e. to natively compile the "graph_lenet" example for Linux 64bit:
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	744
Isabella Gottardi	b28f29d	2017-11-09 17:05:07 +0000	[diff] [blame]	745	g++ examples/graph_lenet.cpp utils/Utils.cpp utils/GraphUtils.cpp -I. -Iinclude -std=c++11 L. -larm_compute_graph -larm_compute -larm_compute_core -Wl,--allow-shlib-undefined -o graph_lenet
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	746
				747	(notice the only difference with the 32 bit command is that we don't need the -mfpu option)
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	748
Anthony Barbier	e500747	2017-10-27 15:01:44 +0100	[diff] [blame]	749	@note If compiling using static libraries, this order must be followed when linking: arm_compute_graph_static, arm_compute, arm_compute_core
				750
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	751	@note These two commands assume libarm_compute.so is available in your library path, if not add the path to it using -L
				752
				753	To run the built executable simply run:
				754
				755	LD_LIBRARY_PATH=build ./neon_convolution
				756
				757	or
				758
				759	LD_LIBRARY_PATH=build ./cl_convolution
				760
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	761	@note Examples accept different types of arguments, to find out what they are run the example without any argument and the help will be displayed at the beginning of the run.
				762
				763	For example:
				764	LD_LIBRARY_PATH=. ./graph_lenet
				765
				766	./graph_lenet
				767
				768	Usage: ./graph_lenet [target] [path_to_data] [batches]
				769
				770	No data folder provided: using random values
				771
				772	Test passed
				773
				774	In this case the first argument of LeNet (like all the graph examples) is the target (i.e 0 to run on NEON, 1 to run on OpenCL if available, 2 to run on OpenCL using the CLTuner), the second argument is the path to the folder containing the npy files for the weights and finally the third argument is the number of batches to run.
				775
Moritz Pflanzer	07674de	2017-07-21 09:39:36 +0100	[diff] [blame]	776	@subsection S3_3_android Building for Android
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	777
				778	For Android, the library was successfully built and tested using Google's standalone toolchains:
Anthony Barbier	a8a28f6	2018-02-26 19:16:32 +0000	[diff] [blame]	779	- clang++ from NDK r16b for armv7a
				780	- clang++ from NDK r16b for arm64-v8a
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	781
				782	Here is a guide to <a href="https://developer.android.com/ndk/guides/standalone_toolchain.html">create your Android standalone toolchains from the NDK</a>
				783
Anthony Barbier	a8a28f6	2018-02-26 19:16:32 +0000	[diff] [blame]	784	- Download the NDK r16b from here: https://developer.android.com/ndk/downloads/index.html
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	785	- Make sure you have Python 2 installed on your machine.
				786	- Generate the 32 and/or 64 toolchains by running the following commands:
				787
Anthony Barbier	a8a28f6	2018-02-26 19:16:32 +0000	[diff] [blame]	788	$NDK/build/tools/make_standalone_toolchain.py --arch arm64 --install-dir $MY_TOOLCHAINS/aarch64-linux-android-ndk-r16b --stl gnustl --api 21
				789	$NDK/build/tools/make_standalone_toolchain.py --arch arm --install-dir $MY_TOOLCHAINS/arm-linux-android-ndk-r16b --stl gnustl --api 21
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	790
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	791	@attention Due to some NDK issues make sure you use clang++ & gnustl
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	792
				793	@note Make sure to add the toolchains to your PATH: export PATH=$PATH:$MY_TOOLCHAINS/aarch64-linux-android-4.9/bin:$MY_TOOLCHAINS/arm-linux-androideabi-4.9/bin
				794
				795	@subsubsection S3_3_1_library How to build the library ?
				796
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	797	To cross-compile the library in debug mode, with NEON only support, for Android 32bit:
				798
				799	CXX=clang++ CC=clang scons Werror=1 -j8 debug=1 neon=1 opencl=0 os=android arch=armv7a
				800
				801	To cross-compile the library in asserts mode, with OpenCL only support, for Android 64bit:
				802
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	803	CXX=clang++ CC=clang scons Werror=1 -j8 debug=0 asserts=1 neon=0 opencl=1 embed_kernels=1 os=android arch=arm64-v8a
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	804
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	805	To cross-compile the library in asserts mode, with GLES_COMPUTE only support, for Android 64bit:
				806
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	807	CXX=clang++ CC=clang scons Werror=1 -j8 debug=0 asserts=1 neon=0 opencl=0 gles_compute=1 embed_kernels=1 os=android arch=arm64-v8a
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	808
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	809	@subsubsection S3_3_2_examples How to manually build the examples ?
				810
				811	The examples get automatically built by scons as part of the build process of the library described above. This section just describes how you can build and link your own application against our library.
				812
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	813	@note The following command lines assume the arm_compute binaries are present in the current directory or in the system library path. If this is not the case you can specify the location of the pre-built library with the compiler option -L. When building the OpenCL example the commands below assume that the CL headers are located in the include folder where the command is executed.
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	814
				815	Once you've got your Android standalone toolchain built and added to your path you can do the following:
				816
				817	To cross compile a NEON example:
				818
				819	#32 bit:
Georgios Pinitas	9873ea3	2017-12-05 15:28:55 +0000	[diff] [blame]	820	arm-linux-androideabi-clang++ examples/neon_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute-static -larm_compute_core-static -L. -o neon_convolution_arm -static-libstdc++ -pie
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	821	#64 bit:
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	822	aarch64-linux-android-clang++ examples/neon_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute-static -larm_compute_core-static -L. -o neon_convolution_aarch64 -static-libstdc++ -pie
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	823
				824	To cross compile an OpenCL example:
				825
				826	#32 bit:
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	827	arm-linux-androideabi-clang++ examples/cl_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute-static -larm_compute_core-static -L. -o cl_convolution_arm -static-libstdc++ -pie -DARM_COMPUTE_CL
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	828	#64 bit:
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	829	aarch64-linux-android-clang++ examples/cl_convolution.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute-static -larm_compute_core-static -L. -o cl_convolution_aarch64 -static-libstdc++ -pie -DARM_COMPUTE_CL
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	830
				831	To cross compile a GLES example:
Anthony Barbier	cc0a80b	2017-12-15 11:37:29 +0000	[diff] [blame]	832
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	833	#32 bit:
				834	arm-linux-androideabi-clang++ examples/gc_absdiff.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute-static -larm_compute_core-static -L. -o gc_absdiff_arm -static-libstdc++ -pie -DARM_COMPUTE_GC
				835	#64 bit:
				836	aarch64-linux-android-clang++ examples/gc_absdiff.cpp utils/Utils.cpp -I. -Iinclude -std=c++11 -larm_compute-static -larm_compute_core-static -L. -o gc_absdiff_aarch64 -static-libstdc++ -pie -DARM_COMPUTE_GC
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	837
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	838	To cross compile the examples with the Graph API, such as graph_lenet.cpp, you need to link the library arm_compute_graph also.
				839	(notice the compute library has to be built with both neon and opencl enabled - neon=1 and opencl=1)
				840
				841	#32 bit:
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	842	arm-linux-androideabi-clang++ examples/graph_lenet.cpp utils/Utils.cpp utils/GraphUtils.cpp -I. -Iinclude -std=c++11 -Wl,--whole-archive -larm_compute_graph-static -Wl,--no-whole-archive -larm_compute-static -larm_compute_core-static -L. -o graph_lenet_arm -static-libstdc++ -pie -DARM_COMPUTE_CL
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	843	#64 bit:
Georgios Pinitas	d9eb275	2018-04-03 13:44:29 +0100	[diff] [blame^]	844	aarch64-linux-android-clang++ examples/graph_lenet.cpp utils/Utils.cpp utils/GraphUtils.cpp -I. -Iinclude -std=c++11 -Wl,--whole-archive -larm_compute_graph-static -Wl,--no-whole-archive -larm_compute-static -larm_compute_core-static -L. -o graph_lenet_aarch64 -static-libstdc++ -pie -DARM_COMPUTE_CL
Gian Marco Iodice	daec1aa	2017-09-29 12:03:18 +0100	[diff] [blame]	845
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	846	@note Due to some issues in older versions of the Mali OpenCL DDK (<= r13p0), we recommend to link arm_compute statically on Android.
Anthony Barbier	20dbb82	2017-12-13 21:19:39 +0000	[diff] [blame]	847	@note When linked statically the arm_compute_graph library currently needs the --whole-archive linker flag in order to work properly
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	848
				849	Then you need to do is upload the executable and the shared library to the device using ADB:
				850
				851	adb push neon_convolution_arm /data/local/tmp/
				852	adb push cl_convolution_arm /data/local/tmp/
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	853	adb push gc_absdiff_arm /data/local/tmp/
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	854	adb shell chmod 777 -R /data/local/tmp/
				855
				856	And finally to run the example:
				857
				858	adb shell /data/local/tmp/neon_convolution_arm
				859	adb shell /data/local/tmp/cl_convolution_arm
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	860	adb shell /data/local/tmp/gc_absdiff_arm
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	861
				862	For 64bit:
				863
				864	adb push neon_convolution_aarch64 /data/local/tmp/
				865	adb push cl_convolution_aarch64 /data/local/tmp/
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	866	adb push gc_absdiff_aarch64 /data/local/tmp/
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	867	adb shell chmod 777 -R /data/local/tmp/
				868
				869	And finally to run the example:
				870
				871	adb shell /data/local/tmp/neon_convolution_aarch64
				872	adb shell /data/local/tmp/cl_convolution_aarch64
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	873	adb shell /data/local/tmp/gc_absdiff_aarch64
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	874
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	875	@note Examples accept different types of arguments, to find out what they are run the example without any argument and the help will be displayed at the beginning of the run.
				876
				877	For example:
				878	adb shell /data/local/tmp/graph_lenet
				879
				880	/data/local/tmp/graph_lenet
				881
				882	Usage: /data/local/tmp/graph_lenet [target] [path_to_data] [batches]
				883
				884	No data folder provided: using random values
				885
				886	Test passed
				887
				888	In this case the first argument of LeNet (like all the graph examples) is the target (i.e 0 to run on NEON, 1 to run on OpenCL if available, 2 to run on OpenCL using the CLTuner), the second argument is the path to the folder containing the npy files for the weights and finally the third argument is the number of batches to run.
				889
Michalis Spyrou	6e52ba3	2017-10-04 15:40:38 +0100	[diff] [blame]	890	@subsection S3_4_bare_metal Building for bare metal
				891
				892	For bare metal, the library was successfully built using linaros's latest (gcc-linaro-6.3.1-2017.05) bare metal toolchains:
				893	- arm-eabi for armv7a
				894	- aarch64-elf for arm64-v8a
				895
				896	Download linaro for <a href="https://releases.linaro.org/components/toolchain/binaries/6.3-2017.05/arm-eabi/">armv7a</a> and <a href="https://releases.linaro.org/components/toolchain/binaries/6.3-2017.05/aarch64-elf/">arm64-v8a</a>.
				897
				898	@note Make sure to add the toolchains to your PATH: export PATH=$PATH:$MY_TOOLCHAINS/gcc-linaro-6.3.1-2017.05-x86_64_aarch64-elf/bin:$MY_TOOLCHAINS/gcc-linaro-6.3.1-2017.05-x86_64_arm-eabi/bin
				899
				900	@subsubsection S3_4_1_library How to build the library ?
				901
				902	To cross-compile the library with NEON support for baremetal arm64-v8a:
				903
				904	scons Werror=1 -j8 debug=0 neon=1 opencl=0 os=bare_metal arch=arm64-v8a build=cross_compile cppthreads=0 openmp=0 standalone=1
				905
				906	@subsubsection S3_4_2_examples How to manually build the examples ?
				907
				908	Examples are disabled when building for bare metal. If you want to build the examples you need to provide a custom bootcode depending on the target architecture and link against the compute library. More information about bare metal bootcode can be found <a href="http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dai0527a/index.html">here</a>.
				909
				910	@subsection S3_5_windows_host Building on a Windows host system
Moritz Pflanzer	07674de	2017-07-21 09:39:36 +0100	[diff] [blame]	911
				912	Using `scons` directly from the Windows command line is known to cause
				913	problems. The reason seems to be that if `scons` is setup for cross-compilation
				914	it gets confused about Windows style paths (using backslashes). Thus it is
				915	recommended to follow one of the options outlined below.
				916
Michalis Spyrou	6e52ba3	2017-10-04 15:40:38 +0100	[diff] [blame]	917	@subsubsection S3_5_1_ubuntu_on_windows Bash on Ubuntu on Windows
Moritz Pflanzer	07674de	2017-07-21 09:39:36 +0100	[diff] [blame]	918
				919	The best and easiest option is to use
				920	<a href="https://msdn.microsoft.com/en-gb/commandline/wsl/about">Ubuntu on Windows</a>.
				921	This feature is still marked as beta and thus might not be available.
				922	However, if it is building the library is as simple as opening a *Bash on
				923	Ubuntu on Windows* shell and following the general guidelines given above.
				924
Michalis Spyrou	6e52ba3	2017-10-04 15:40:38 +0100	[diff] [blame]	925	@subsubsection S3_5_2_cygwin Cygwin
Moritz Pflanzer	07674de	2017-07-21 09:39:36 +0100	[diff] [blame]	926
				927	If the Windows subsystem for Linux is not available <a href="https://www.cygwin.com/">Cygwin</a>
				928	can be used to install and run `scons`. In addition to the default packages
				929	installed by Cygwin `scons` has to be selected in the installer. (`git` might
				930	also be useful but is not strictly required if you already have got the source
				931	code of the library.) Linaro provides pre-built versions of
				932	<a href="http://releases.linaro.org/components/toolchain/binaries/">GCC cross-compilers</a>
				933	that can be used from the Cygwin terminal. When building for Android the
				934	compiler is included in the Android standalone toolchain. After everything has
				935	been set up in the Cygwin terminal the general guide on building the library
				936	can be followed.
				937
Michalis Spyrou	6e52ba3	2017-10-04 15:40:38 +0100	[diff] [blame]	938	@subsection S3_6_cl_stub_library The OpenCL stub library
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	939
				940	In the opencl-1.2-stubs folder you will find the sources to build a stub OpenCL library which then can be used to link your application or arm_compute against.
				941
				942	If you preferred you could retrieve the OpenCL library from your device and link against this one but often this library will have dependencies on a range of system libraries forcing you to link your application against those too even though it is not using them.
				943
				944	@warning This OpenCL library provided is a stub and not a real implementation. You can use it to resolve OpenCL's symbols in arm_compute while building the example but you must make sure the real libOpenCL.so is in your PATH when running the example or it will not work.
				945
				946	To cross-compile the stub OpenCL library simply run:
				947
				948	<target-prefix>-gcc -o libOpenCL.so -Iinclude opencl-1.2-stubs/opencl_stubs.c -fPIC -shared
				949
				950	For example:
				951
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	952	#Linux 32bit
				953	arm-linux-gnueabihf-gcc -o libOpenCL.so -Iinclude opencl-1.2-stubs/opencl_stubs.c -fPIC -shared
				954	#Linux 64bit
				955	aarch64-linux-gnu-gcc -o libOpenCL.so -Iinclude -shared opencl-1.2-stubs/opencl_stubs.c -fPIC
				956	#Android 32bit
				957	arm-linux-androideabi-clang -o libOpenCL.so -Iinclude -shared opencl-1.2-stubs/opencl_stubs.c -fPIC -shared
				958	#Android 64bit
Anthony Barbier	14c86a9	2017-12-14 16:27:41 +0000	[diff] [blame]	959	aarch64-linux-android-clang -o libOpenCL.so -Iinclude -shared opencl-1.2-stubs/opencl_stubs.c -fPIC -shared
				960
				961	@subsection S3_7_gles_stub_library The Linux OpenGLES and EGL stub libraries
				962
				963	In the opengles-3.1-stubs folder you will find the sources to build stub EGL and OpenGLES libraries which then can be used to link your Linux application of arm_compute against.
				964
				965	@note The stub libraries are only needed on Linux. For Android, the NDK toolchains already provide the meta-EGL and meta-GLES libraries.
				966
				967	To cross-compile the stub OpenGLES and EGL libraries simply run:
				968
				969	<target-prefix>-gcc -o libEGL.so -Iinclude/linux opengles-3.1-stubs/EGL.c -fPIC -shared
				970	<target-prefix>-gcc -o libGLESv2.so -Iinclude/linux opengles-3.1-stubs/GLESv2.c -fPIC -shared
				971
				972	#Linux 32bit
				973	arm-linux-gnueabihf-gcc -o libEGL.so -Iinclude/linux opengles-3.1-stubs/EGL.c -fPIC -shared
				974	arm-linux-gnueabihf-gcc -o libGLESv2.so -Iinclude/linux opengles-3.1-stubs/GLESv2.c -fPIC -shared
				975
				976	#Linux 64bit
				977	aarch64-linux-gnu-gcc -o libEGL.so -Iinclude/linux opengles-3.1-stubs/EGL.c -fPIC -shared
				978	aarch64-linux-gnu-gcc -o libGLESv2.so -Iinclude/linux opengles-3.1-stubs/GLESv2.c -fPIC -shared
Anthony Barbier	6ff3b19	2017-09-04 18:44:23 +0100	[diff] [blame]	979	*/
Anthony Barbier	3762e74	2018-03-02 11:49:33 +0000	[diff] [blame]	980	} // namespace arm_compute