[BYOC][ACL] Improve installation tutorial

Improves installation script so that ACL can be built natively and improves tutorial to give clearer information on how ACL can be installed using two different methods.

Change-Id: I6cec98b4b0a7dc2b151b36583d3d28f2b85f8702

cmake/modules/contrib/ArmComputeLib.cmake

@@ -36,6 +36,8 @@ if(USE_ARM_COMPUTE_LIB_GRAPH_RUNTIME)
 
     file(GLOB ACL_CONTRIB_SRC src/runtime/contrib/arm_compute_lib/*)
 
+    # Cmake needs to find arm_compute, include and support directories
+    # in the path specified by ACL_PATH.
     set(ACL_INCLUDE_DIRS ${ACL_PATH}/include ${ACL_PATH})
     include_directories(${ACL_INCLUDE_DIRS})
 

docker/install/ubuntu_install_arm_compute_lib.sh

@@ -24,7 +24,7 @@ repo_url="https://github.com/ARM-software/ComputeLibrary.git"
 repo_dir="acl"
 install_path="/opt/$repo_dir"
 architecture_type=$(uname -i)
-target_arch="arm64-v8a" # arm64-v8a/armv7a
+target_arch="arm64-v8a" # arm64-v8a / arm64-v8.2-a / armv7a
 build_type="native"
 
 tmpdir=$(mktemp -d)
@@ -41,9 +41,16 @@ apt-get install -y --no-install-recommends \
     git \
     scons \
     bsdmainutils \
-    build-essential \
+    build-essential
+
+# Install cross-compiler when not building natively.
+# Depending on the architecture selected to compile for,
+# you may need to install an alternative cross-compiler.
+if [ "$architecture_type" != "aarch64" ]; then
+  apt-get install -y --no-install-recommends \
     g++-aarch64-linux-gnu \
     gcc-aarch64-linux-gnu
+fi
 
 cd "$tmpdir"
 

docs/deploy/arm_compute_lib.rst

@@ -26,6 +26,24 @@ and GPU's. Currently the integration offloads operators to ACL to use hand-craft
 routines in the library. By offloading select operators from a relay graph to ACL we can achieve
 a performance boost on such devices.
 
+Installing Arm Compute Library
+------------------------------
+
+We recommend two different ways to build and install ACL:
+
+* Use the script located at `docker/install/ubuntu_install_arm_compute_library.sh`. You can use this
+  script for building ACL from source natively or for cross-compiling the library on an x86 machine.
+  You may need to change the architecture of the device you wish to compile for by altering the
+  `target_arch` variable. Binaries will be built from source and installed to the location denoted by
+  `install_path`.
+* Alternatively, you can download and use pre-built binaries from:
+  https://github.com/ARM-software/ComputeLibrary/releases. When using this package, you will need to
+  select the binaries for the architecture you require and make sure they are visible to cmake.
+
+In both cases you will need to set USE_ARM_COMPUTE_LIB_GRAPH_RUNTIME to the path where the ACL package
+is located. Cmake will look in /path-to-acl/ along with /path-to-acl/lib and /path-to-acl/build for the
+required binaries. See the section below for more information on how to use these configuration options.
+
 Building with ACL support
 -------------------------
 
@@ -42,6 +60,19 @@ to compile an ACL module on an x86 machine and then run the module on a remote A
 need to use USE_ARM_COMPUTE_LIB=ON on the x86 machine and USE_ARM_COMPUTE_LIB_GRAPH_RUNTIME=ON on the remote
 AArch64 device.
 
+Using USE_ARM_COMPUTE_LIB_GRAPH_RUNTIME=ON will mean that ACL binaries are searched for by cmake in the
+default locations (see https://cmake.org/cmake/help/v3.4/command/find_library.html). In addition to this,
+/path-to-tvm-project/acl/ will also be searched. It is likely that you will need to set your own path to
+locate ACL. This can be done by specifying a path in the place of ON.
+
+These flags should be set in your config.cmake file. For example:
+
+.. code:: cmake
+
+    set(USE_ARM_COMPUTE_LIB ON)
+    set(USE_ARM_COMPUTE_LIB_GRAPH_RUNTIME /path/to/acl)
+
+
 Usage
 -----
 
@@ -160,7 +191,7 @@ There are a series of files we need to make changes to:
 offloadable.
 * `src/relay/backend/contrib/arm_compute_lib/codegen.cc` Implement `Create[OpName]JSONNode` method. This is where we
 declare how the operator should be represented by JSON. This will be used to create the ACL module.
-* `src/runtime/contrib/arm_compute_lib/acl_kernel.h` Implement `Create[OpName]Layer` method. This is where we
+* `src/runtime/contrib/arm_compute_lib/acl_runtime.cc` Implement `Create[OpName]Layer` method. This is where we
 define how the JSON representation can be used to create an ACL function. We simply define how to
 translate from the JSON representation to ACL API.
 * `tests/python/contrib/test_arm_compute_lib` Add unit tests for the given operator.