COMPMID-3869: Update Sconstruct to support SVE/SVE2

Modifying scons to build with SVE/SVE2
Updating the documentation with examples

Change-Id: I80875206599d5444b9c21ac75c4a8e4efd30d8b5
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4629
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Sang-Hoon Park <sang-hoon.park@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>

4 files changed, 174 insertions, 71 deletions

diff --git a/SConstruct b/SConstruct
index 395fb5e59d..e9ba496afb 100644
--- a/SConstruct
+++ b/SConstruct
@@ -43,8 +43,8 @@ vars.AddVariables(
     BoolVariable("asserts", "Enable asserts (this flag is forced to 1 for debug=1)", False),
     BoolVariable("logging", "Logging (this flag is forced to 1 for debug=1)", False),
     EnumVariable("arch", "Target Architecture", "armv7a",
-                  allowed_values=("armv7a", "arm64-v8a", "arm64-v8.2-a", "arm64-v8.2-a-sve", "x86_32", "x86_64",
-                                  "armv8a", "armv8.2-a", "armv8.2-a-sve", "armv8.6-a", "armv8.6-a-sve", "x86")),
+                  allowed_values=("armv7a", "arm64-v8a", "arm64-v8.2-a", "arm64-v8.2-a-sve", "arm64-v8.2-a-sve2", "x86_32", "x86_64",
+                                  "armv8a", "armv8.2-a", "armv8.2-a-sve", "armv8.6-a", "armv8.6-a-sve", "armv8.6-a-sve2", "x86")),
     EnumVariable("estate", "Execution State", "auto", allowed_values=("auto", "32", "64")),
     EnumVariable("os", "Target OS", "linux", allowed_values=("linux", "android", "tizen", "bare_metal")),
     EnumVariable("build", "Build type", "cross_compile", allowed_values=("native", "cross_compile", "embed_only")),
@@ -72,7 +72,8 @@ vars.AddVariables(
     ("compiler_prefix", "Override the compiler prefix", ""),
     ("extra_cxx_flags", "Extra CXX flags to be appended to the build command", ""),
     ("extra_link_flags", "Extra LD flags to be appended to the build command", ""),
-    ("compiler_cache", "Command to prefix to the C and C++ compiler (e.g ccache)", "")
+    ("compiler_cache", "Command to prefix to the C and C++ compiler (e.g ccache)", ""),
+    ("specs_file", "Specs file to use (e.g. rdimon.specs)", "")
 )
 
 env = Environment(platform="posix", variables=vars, ENV = os.environ)
@@ -202,7 +203,9 @@ if 'v7a' in env['arch']:
     else:
         env.Append(CXXFLAGS = ['-mfloat-abi=hard'])
 elif 'v8' in env['arch']:
-    if 'sve' in env['arch']:
+    if 'sve2' in env['arch']:
+        env.Append(CXXFLAGS = ['-march=armv8.2-a+sve2+fp16+dotprod'])
+    elif 'sve' in env['arch']:
         env.Append(CXXFLAGS = ['-march=armv8.2-a+sve+fp16+dotprod'])
     elif 'v8.2-a' in env['arch']:
         env.Append(CXXFLAGS = ['-march=armv8.2-a+fp16']) # explicitly enable fp16 extension otherwise __ARM_FEATURE_FP16_VECTOR_ARITHMETIC is undefined
@@ -312,13 +315,17 @@ if env['os'] == 'android':
     env.Append(LINKFLAGS = ['-pie', '-static-libstdc++', '-ldl'])
 elif env['os'] == 'bare_metal':
     env.Append(LINKFLAGS = ['-static'])
-    env.Append(LINKFLAGS = ['-specs=rdimon.specs'])
     env.Append(CXXFLAGS = ['-fPIC'])
+    if env['specs_file'] == "":
+        env.Append(LINKFLAGS = ['-specs=rdimon.specs'])
     env.Append(CPPDEFINES = ['NO_MULTI_THREADING'])
     env.Append(CPPDEFINES = ['BARE_METAL'])
 if env['os'] == 'linux' and env['arch'] == 'armv7a':
     env.Append(CXXFLAGS = [ '-Wno-psabi' ])
 
+if env['specs_file'] != "":
+    env.Append(LINKFLAGS = ['-specs='+env['specs_file']])
+
 if env['opencl']:
     if env['os'] in ['bare_metal'] or env['standalone']:
         print("Cannot link OpenCL statically, which is required for bare metal / standalone builds")
diff --git a/docs/00_introduction.dox b/docs/00_introduction.dox
index 94f73f8cdb..49ccbce5c3 100644
--- a/docs/00_introduction.dox
+++ b/docs/00_introduction.dox
@@ -1309,85 +1309,169 @@ v16.12 Binary preview release
 scons 2.3 or above is required to build the library.
 To see the build options available simply run ```scons -h```:
 
-	debug: Debug (yes|no)
-		default: False
-		actual: False
+        debug: Debug (yes|no)
+            default: False
+            actual: False
+
+        asserts: Enable asserts (this flag is forced to 1 for debug=1) (yes|no)
+            default: False
+            actual: False
+
+        logging: Logging (this flag is forced to 1 for debug=1) (yes|no)
+            default: False
+            actual: False
+
+        arch: Target Architecture (armv7a|arm64-v8a|arm64-v8.2-a|arm64-v8.2-a-sve|arm64-v8.2-a-sve2|x86_32|x86_64|armv8a|armv8.2-a|armv8.2-a-sve|armv8.6-a|armv8.6-a-sve|armv8.6-a-sve2|x86)
+            default: armv7a
+            actual: armv7a
+
+        estate: Execution State (auto|32|64)
+            default: auto
+            actual: auto
+
+        os: Target OS (linux|android|tizen|bare_metal)
+            default: linux
+            actual: linux
+
+        build: Build type (native|cross_compile|embed_only)
+            default: cross_compile
+            actual: cross_compile
+
+        examples: Build example programs (yes|no)
+            default: True
+            actual: True
+
+        gemm_tuner: Build gemm_tuner programs (yes|no)
+            default: True
+            actual: True
+
+        Werror: Enable/disable the -Werror compilation flag (yes|no)
+            default: True
+            actual: True
+
+        standalone: Builds the tests as standalone executables, links statically with libgcc, libstdc++ and libarm_compute (yes|no)
+            default: False
+            actual: False
+
+        opencl: Enable OpenCL support (yes|no)
+            default: True
+            actual: True
+
+        neon: Enable Neon support (yes|no)
+            default: False
+            actual: False
+
+        gles_compute: Enable OpenGL ES Compute Shader support (yes|no)
+            default: False
+            actual: False
+
+        embed_kernels: Embed OpenCL kernels and OpenGL ES compute shaders in library binary (yes|no)
+            default: True
+            actual: True
+
+        set_soname: Set the library's soname and shlibversion (requires SCons 2.4 or above) (yes|no)
+            default: False
+            actual: False
+
+        tracing: Enable runtime tracing (yes|no)
+            default: False
+            actual: False
+
+        openmp: Enable OpenMP backend (yes|no)
+            default: False
+            actual: False
+
+        cppthreads: Enable C++11 threads backend (yes|no)
+            default: True
+            actual: True
+
+        build_dir: Specify sub-folder for the build ( /path/to/build_dir )
+            default: .
+            actual: .
+
+        install_dir: Specify sub-folder for the install ( /path/to/install_dir )
+            default:
+            actual:
+
+        exceptions: Enable/disable C++ exception support (yes|no)
+            default: True
+            actual: True
+
+        linker_script: Use an external linker script ( /path/to/linker_script )
+            default:
+            actual:
+
+        internal_only: Enable ARM internal only tests (yes|no)
+            default: False
+            actual: False
+
+        custom_options: Custom options that can be used to turn on/off features
+            (all|none|comma-separated list of names)
+            allowed names: disable_mmla_fp
+            default: none
+            actual:
 
-	asserts: Enable asserts (this flag is forced to 1 for debug=1) (yes|no)
-		default: False
-		actual: False
+        data_type_support: Enable a list of data types to support
+            (all|none|comma-separated list of names)
+            allowed names: qasymm8 qasymm8_signed qsymm16 fp16 fp32
+            default: all
+            actual: qasymm8 qasymm8_signed qsymm16 fp16 fp32
 
-	arch: Target Architecture (armv7a|arm64-v8a|arm64-v8.2-a|x86_32|x86_64)
-		default: armv7a
-		actual: armv7a
+        toolchain_prefix: Override the toolchain prefix
+            default:
+            actual:
 
-	os: Target OS (linux|android|bare_metal)
-		default: linux
-		actual: linux
+        compiler_prefix: Override the compiler prefix
+            default:
+            actual:
 
-	build: Build type (native|cross_compile|embed_only)
-		default: cross_compile
-		actual: cross_compile
+        extra_cxx_flags: Extra CXX flags to be appended to the build command
+            default:
+            actual:
 
-	examples: Build example programs (yes|no)
-		default: True
-		actual: True
+        extra_link_flags: Extra LD flags to be appended to the build command
+            default:
+            actual:
 
-	Werror: Enable/disable the -Werror compilation flag (yes|no)
-		default: True
-		actual: True
+        compiler_cache: Command to prefix to the C and C++ compiler (e.g ccache)
+            default:
+            actual:
 
-	opencl: Enable OpenCL support (yes|no)
-		default: True
-		actual: True
+        specs_file: Specs file to use
+            default: rdimon.specs
+            actual: rdimon.specs
 
-	neon: Enable Neon support (yes|no)
-		default: False
-		actual: False
+        benchmark_examples: Build benchmark examples programs (yes|no)
+            default: True
+            actual: True
 
-	gles_compute: Enable OpenGL ES Compute Shader support (yes|no)
-		default: False
-		actual: False
+        validate_examples: Build validate examples programs (yes|no)
+            default: True
+            actual: True
 
-	embed_kernels: Embed OpenCL kernels and OpenGL ES compute shader in library binary (yes|no)
-		default: True
-		actual: True
+        reference_openmp: Build reference validation with openmp (yes|no)
+            default: True
+            actual: True
 
-	set_soname: Set the library's soname and shlibversion (requires SCons 2.4 or above) (yes|no)
-		default: False
-		actual: False
+        validation_tests: Build validation test programs (yes|no)
+            default: True
+            actual: True
 
-	openmp: Enable OpenMP backend (yes|no)
-		default: False
-		actual: False
+        benchmark_tests: Build benchmark test programs (yes|no)
+            default: True
+            actual: True
 
-	cppthreads: Enable C++11 threads backend (yes|no)
-		default: True
-		actual: True
+        test_filter: Pattern to specify the tests' filenames to be compiled
+            default: *.cpp
+            actual: *.cpp
 
-	build_dir: Specify sub-folder for the build ( /path/to/build_dir )
-		default: .
-		actual: .
+        pmu: Enable PMU counters (yes|no)
+            default: False
+            actual: False
 
-	extra_cxx_flags: Extra CXX flags to be appended to the build command
-		default:
-		actual:
-
-	pmu: Enable PMU counters (yes|no)
-		default: False
-		actual: False
-
-	mali: Enable Mali hardware counters (yes|no)
-		default: False
-		actual: False
-
-	validation_tests: Build validation test programs (yes|no)
-		default: False
-		actual: False
-
-	benchmark_tests: Build benchmark test programs (yes|no)
-		default: False
-		actual: False
+        mali: Enable Mali hardware counters (yes|no)
+            default: False
+            actual: False
 
 @b debug / @b asserts:
  - With debug=1 asserts are enabled, and the library is built with symbols and no optimisations enabled.
@@ -1583,6 +1667,18 @@ For example:
 Below is a list of the common parameters among the graph examples :
 @snippet utils/CommonGraphOptions.h Common graph examples parameters
 
+@subsubsection S3_2_3_sve Build for SVE or SVE2
+
+In order to build for SVE or SVE2 you need a compiler that supports them. You can find more information in the following these links:
+    -# GCC: https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/sve-support
+    -# LLVM: https://developer.arm.com/tools-and-software/open-source-software/developer-tools/llvm-toolchain/sve-support
+
+@note You the need to indicate the toolchains using the scons "toolchain_prefix" parameter.
+
+An example build command with SVE is:
+
+        scons arch=arm64-v8.2-a-sve os=linux build_dir=arm64 -j55 standalone=0 opencl=0 openmp=0 validation_tests=1 neon=1 cppthreads=1 toolchain_prefix=aarch64-none-linux-gnu-
+
 @subsection S3_3_android Building for Android
 
 For Android, the library was successfully built and tested using Google's standalone toolchains:
diff --git a/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_s8s32_dot_8x1VL/generic.cpp b/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_s8s32_dot_8x1VL/generic.cpp
index 5770076d04..98004e98a5 100644
--- a/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_s8s32_dot_8x1VL/generic.cpp
+++ b/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_s8s32_dot_8x1VL/generic.cpp
@@ -33,7 +33,7 @@
 
 namespace arm_gemm {
 
-void sve_smallK_hybrid_s8s32_dot_8x1VL(const int8_t *A, int lda, const int8_t *B, int32_t *C, int ldc, int M, int N, int K, const int32_t *bias, Activation act, bool) {
+void sve_smallK_hybrid_s8s32_dot_8x1VL(const int8_t *A, int lda, const int8_t *B, int32_t *C, int ldc, int M, int N, int K, const int32_t *, Activation, bool) {
     const long loops_count = iceildiv(N, (int)get_vector_length<int32_t>()) - 1;
     const long ldab = lda * sizeof(int8_t);
     const long ldcb = ldc * sizeof(int32_t);
diff --git a/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_u8u32_dot_8x1VL/generic.cpp b/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_u8u32_dot_8x1VL/generic.cpp
index b980d9b5c2..6a8553216b 100644
--- a/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_u8u32_dot_8x1VL/generic.cpp
+++ b/src/core/NEON/kernels/arm_gemm/kernels/sve_smallK_hybrid_u8u32_dot_8x1VL/generic.cpp
@@ -33,7 +33,7 @@
 
 namespace arm_gemm {
 
-void sve_smallK_hybrid_u8u32_dot_8x1VL(const uint8_t *A, int lda, const uint8_t *B, uint32_t *C, int ldc, int M, int N, int K, const uint32_t *bias, Activation act, bool) {
+void sve_smallK_hybrid_u8u32_dot_8x1VL(const uint8_t *A, int lda, const uint8_t *B, uint32_t *C, int ldc, int M, int N, int K, const uint32_t *, Activation , bool) {
     const long loops_count = iceildiv(N, (int)get_vector_length<uint32_t>()) - 1;
     const long ldab = lda * sizeof(uint8_t);
     const long ldcb = ldc * sizeof(uint32_t);