In this document
Related Samples
Renderscript exposes a set of compute APIs that you can use to do intensive computational operations. You can use the compute APIs in the context of a graphics Renderscript such as calculating the positions of many objects in a scene. You can also create standalone compute Renderscripts such as one that does image processing for a photo editor application.
Compute Renderscripts scale to the amount of
processing cores available on the device. This is enabled through a function named
rsForEach()
(or the forEach_root()
method at the Android framework level).
that automatically partitions work across available processing cores on the device.
For now, compute Renderscripts can only take advantage of CPU
cores, but in the future, they can potentially run on other types of processors such as GPUs and
DSPs.
Creating a Compute Renderscript
Implementing a compute Renderscript creating a .rs
file that contains
your Renderscript code and calling it at the Android framework level with the
forEach_root()
or at the Renderscript runtime level with the
rsForEach()
function. The following diagram describes how a typical compute
Renderscript is set up:
The following sections describe how to create a simple compute Renderscript and use it in an Android application. This example uses the HelloCompute Renderscript sample that is provided in the SDK as a guide (some code has been modified from its original form for simplicity).
Creating the Renderscript file
Your Renderscript code resides in .rs
and .rsh
files in the
<project_root>/src/
directory. This code contains the compute logic
and declares all necessary variables and pointers.
Every compute .rs
file generally contains the following items:
- A pragma declaration (
#pragma rs java_package_name(package.name)
) that declares the package name of the.java
reflection of this Renderscript. - A pragma declaration (
#pragma version(1)
) that declares the version of Renderscript that you are using (1 is the only value for now). - A
root()
function that is the main worker function. The root function is called by thersForEach
function, which allows the Renderscript code to be called and executed on multiple cores if they are available. Theroot()
function must returnvoid
and accept the following arguments:- Pointers to memory allocations that are used for the input and output of the compute Renderscript. Both of these pointers are required for Android 3.2 (API level 13) platform versions or older. Android 4.0 (API level 14) and later requires one or both of these allocations.
The following arguments are optional, but both must be supplied if you choose to use them:
- A pointer for user-defined data that the Renderscript might need to carry out computations in addition to the necessary allocations. This can be a pointer to a simple primitive or a more complex struct.
- The size of the user-defined data.
- An optional
init()
function. This allows you to do any initialization before theroot()
function runs, such as initializing variables. This function runs once and is called automatically when the Renderscript starts, before anything else in your Renderscript. - Any variables, pointers, and structures that you wish to use in your Renderscript code (can
be declared in
.rsh
files if desired)
The following code shows how the mono.rs file is implemented:
#pragma version(1) #pragma rs java_package_name(com.example.android.rs.hellocompute) //multipliers to convert a RGB colors to black and white const static float3 gMonoMult = {0.299f, 0.587f, 0.114f}; void root(const uchar4 *v_in, uchar4 *v_out) { //unpack a color to a float4 float4 f4 = rsUnpackColor8888(*v_in); //take the dot product of the color and the multiplier float3 mono = dot(f4.rgb, gMonoMult); //repack the float to a color *v_out = rsPackColorTo8888(mono); }
Calling the Renderscript code
You can do Renderscript to Renderscript calls with rsForEach
in situations
such as when a graphics Renderscript needs to do a lot of computational operations. The Renderscript
Balls sample shows how
this is setup. The balls.rs
graphics Renderscript calls the balls_physics.rs
compute Renderscript to calculate the location of the balls that are rendered to the screen.
Another way to use a compute Renderscript is to call it from your Android framework code by
creating a Renderscript object by instantiating the (ScriptC_script_name
)
class. This class contains a method, forEach_root()
, that lets you invoke
rsForEach
. You give it the same parameters that you would if you were invoking it
at the Renderscript runtime level. This technique allows your Android application to offload
intensive mathematical calculations to Renderscript. See the HelloCompute sample to see
how a simple Android application can utilize a compute Renderscript.
To call a compute Renderscript at the Android framework level:
- Allocate memory that is needed by the compute Renderscript in your Android framework code.
You need an input and output
Allocation
for Android 3.2 (API level 13) platform versions and older. The Android 4.0 (API level 14) platform version requires only one or bothAllocation
s. - Create an instance of the
ScriptC_script_name
class. - Call
forEach_root()
, passing in the allocations, the Renderscript, and any optional user-defined data. The output allocation will contain the output of the compute Renderscript.
In the following example, taken from the HelloCompute sample, processes
a bitmap and outputs a black and white version of it. The
createScript()
method carries out the steps described previously. This method the compute
Renderscript, mono.rs
, passing in memory allocations that store the bitmap to be processed
as well as the eventual output bitmap. It then displays the processed bitmap onto the screen:
package com.example.android.rs.hellocompute; import android.app.Activity; import android.os.Bundle; import android.graphics.BitmapFactory; import android.graphics.Bitmap; import android.renderscript.RenderScript; import android.renderscript.Allocation; import android.widget.ImageView; public class HelloCompute extends Activity { private Bitmap mBitmapIn; private Bitmap mBitmapOut; private RenderScript mRS; private Allocation mInAllocation; private Allocation mOutAllocation; private ScriptC_mono mScript; @Override protected void onCreate(Bundle savedInstanceState) { super.onCreate(savedInstanceState); setContentView(R.layout.main); mBitmapIn = loadBitmap(R.drawable.data); mBitmapOut = Bitmap.createBitmap(mBitmapIn.getWidth(), mBitmapIn.getHeight(), mBitmapIn.getConfig()); ImageView in = (ImageView) findViewById(R.id.displayin); in.setImageBitmap(mBitmapIn); ImageView out = (ImageView) findViewById(R.id.displayout); out.setImageBitmap(mBitmapOut); createScript(); } private void createScript() { mRS = RenderScript.create(this); mInAllocation = Allocation.createFromBitmap(mRS, mBitmapIn, Allocation.MipmapControl.MIPMAP_NONE, Allocation.USAGE_SCRIPT); mOutAllocation = Allocation.createTyped(mRS, mInAllocation.getType()); mScript = new ScriptC_mono(mRS, getResources(), R.raw.mono); mScript.forEach_root(mInAllocation, mOutAllocation); mOutAllocation.copyTo(mBitmapOut); } private Bitmap loadBitmap(int resource) { final BitmapFactory.Options options = new BitmapFactory.Options(); options.inPreferredConfig = Bitmap.Config.ARGB_8888; return BitmapFactory.decodeResource(getResources(), resource, options); } }
To call a compute Renderscript from another Renderscript file:
- Allocate memory that is needed by the compute Renderscript in your Android framework code.
You need an input and output
Allocation
for Android 3.2 (API level 13) platform versions and older. The Android 4.0 (API level 14) platform version requires only one or bothAllocation
s. - Call
rsForEach()
, passing in the allocations and any optional user-defined data. The output allocation will contain the output of the compute Renderscript.
The following example, taken from the Renderscript Balls sample, demonstrates how to do make a script to script call:
rs_script script; rs_allocation in_allocation; rs_allocation out_allocation; UserData_t data; ... rsForEach(script, in_allocation, out_allocation, &data, sizeof(data));
In this example, assume that the script and memory allocations have already been
allocated and bound at the Android framework level and that UserData_t
is a struct
declared previously. Passing a pointer to a struct and the size of the struct to rsForEach
is optional, but useful if your compute Renderscript requires additional information other than
the necessary memory allocations.