Core vs Runtime libraries
The Core library is a low level collection of algorithms implementations, it is designed to be embedded in existing projects and applications:
- It doesn't allocate any memory (All the memory allocations/mappings have to be handled by the caller).
- It doesn't perform any kind of multi-threading (but provide information to the caller about how the workload can be split).
The Runtime library is a very basic wrapper around the Core library which can be used for quick prototyping, it is basic in the sense that:
- It allocates images and tensors by using standard malloc().
- It multi-threads Neon code in a very basic way using a very simple pool of threads.
- For OpenCL it uses the default CLScheduler command queue for all mapping operations and kernels.
For maximum performance, it is expected that the users would re-implement an equivalent to the runtime library which suits better their needs (With a more clever multi-threading strategy, load-balancing between Neon and OpenCL, etc.)
Data-type and Data-layout support
Compute Library supports a wide list of data-types, information can been directly found in the documentation of each kernel/function. The main data-types that the Machine Learning functions support are the following:
- BFLOAT16: 16-bit non-standard brain floating point
- F16: 16-bit half precision floating point
- F32: 32-bit single precision floating point
- QASYMM8: 8-bit unsigned asymmetric quantized
- QASYMM8_SIGNED: 8-bit signed asymmetric quantized
- QSYMM8_PER_CHANNEL: 8-bit signed symmetric quantized (Used for the weights)
Moreover, Compute Library supports the following data layouts (fast changing dimension from right to left):
- NHWC: The native layout of Compute Library that delivers the best performance where channels are in the fastest changing dimension
- NCHW: Legacy layout where width is in the fastest changing dimension where N = batches, C = channels, H = height, W = width
Fast-math support
Compute Library supports different types of convolution methods, fast-math flag is only used for the Winograd algorithm. When the fast-math flag is enabled, both Neon and CL convolution layers will try to dispatch the fastest implementation available, which may introduce a drop in accuracy as well. The different scenarios involving the fast-math flag are presented below:
- For FP32:
- no-fast-math: Only supports Winograd 3x3,3x1,1x3,5x1,1x5,7x1,1x7
- fast-math: Supports Winograd 3x3,3x1,1x3,5x1,1x5,7x1,1x7,5x5,7x7
- For fp16:
- no-fast-math: No Winograd support
- fast-math: Supports Winograd 3x3,3x1,1x3,5x1,1x5,7x1,1x7,5x5,7x7
Thread-safety
Although the library supports multi-threading during workload dispatch, thus parallelizing the execution of the workload at multiple threads, the current runtime module implementation is not thread-safe in the sense of executing different functions from separate threads. This lies to the fact that the provided scheduling mechanism wasn't designed with thread-safety in mind. As it is true with the rest of the runtime library a custom scheduling mechanism can be re-implemented to account for thread-safety if needed and be injected as the library's default scheduler.
Windows, kernels, multi-threading and functions
Windows
A Window represents a workload to execute, it can handle up to Coordinates::num_max_dimensions dimensions. Each dimension is defined by a start, end and step.
It can split into subwindows as long as all the following rules remain true for all the dimensions:
Kernels
Each implementation of the IKernel interface (base class of all the kernels in the core library) works in the same way:
OpenCL kernels:
MyKernel kernel;
kernel.configure(
input, output, option0, option1);
const Window& max_window = kernel.window();
kernel.run( q, max_window );
q.finish();
Neon / CPP kernels:
MyKernel kernel;
kernel.configure(
input, output, option0, option1);
const Window& max_window = kernel.window();
kernel.run( max_window );
Multi-threading
The previous section shows how to run a Neon / CPP kernel in the current thread, however if your system has several CPU cores, you will probably want the kernel to use several cores. Here is how this can be done:
info.cpu_info = &_cpu_info;
const Window &max_window = kernel->window();
const unsigned int num_iterations = max_window.num_iterations(split_dimension);
info.num_threads = std::min(num_iterations, _num_threads);
if(num_iterations == 0)
{
return;
}
if(!kernel->is_parallelisable() || info.num_threads == 1)
{
kernel->run(max_window, info);
}
else
{
auto thread_it = _threads.begin();
for(; t < info.num_threads - 1; ++
t, ++thread_it)
{
Window win = max_window.split_window(split_dimension, t, info.num_threads);
thread_it->start(kernel, win, info);
}
Window win = max_window.split_window(split_dimension, t, info.num_threads);
kernel->run(win, info);
try
{
for(auto &thread : _threads)
{
thread.wait();
}
}
catch(const std::system_error &e)
{
std::cerr << "Caught system_error with code " << e.code() << " meaning " << e.what() << '\n';
}
}
This is a very basic implementation which was originally used in the Neon runtime library by all the Neon functions.
- See also
- CPPScheduler
- Note
- Some kernels like for example NEHistogramKernel need some local temporary buffer to perform their calculations. In order to avoid memory corruption between threads, the local buffer must be of size:
memory_needed_per_thread * num_threads and a unique thread_id between 0 and num_threads must be assigned to the ThreadInfo object passed to the run function.
Functions
Functions will automatically allocate the temporary buffers mentioned above, and will automatically multi-thread kernels' executions using the very basic scheduler described in the previous section.
Simple functions only call a single kernel (e.g NEConvolution3x3), while more complex ones consist of several kernels pipelined together (e.g NEGaussianPyramid, NEHarrisCorners). Check their documentation to find out which kernels are used by each function.
MyFunction function;
function.configure(
input, output, option0, option1);
function.run();
- Warning
- The Compute Library requires Mali OpenCL DDK r8p0 or higher (OpenCL kernels are compiled using the -cl-arm-non-uniform-work-group-size flag)
- Note
- All OpenCL functions and objects in the runtime library use the command queue associated with CLScheduler for all operations, a real implementation would be expected to use different queues for mapping operations and kernels in order to reach a better GPU utilization.
OpenCL Scheduler and kernel library
The Compute Library runtime uses a single command queue and context for all the operations.
The user can get / set this context and command queue through CLScheduler's interface.
The user can get / set the target GPU device through the CLScheduler's interface.
- Attention
- Make sure the application is using the same context as the library as in OpenCL it is forbidden to share objects across contexts. This is done by calling CLScheduler::init() or CLScheduler::default_init() at the beginning of your application.
-
Make sure the scheduler's target is not changed after function classes are created.
All OpenCL kernels used by the library are built and stored in CLKernelLibrary. If the library is compiled with embed_kernels=0 the application can set the path to the OpenCL kernels by calling CLKernelLibrary::init(), by default the path is set to "./cl_kernels"
OpenCL events and synchronization
In order to block until all the jobs in the CLScheduler's command queue are done executing the user can call CLScheduler::sync() or create a sync event using CLScheduler::enqueue_sync_event()
For example:
PPMLoader ppm;
constexpr int scale_factor = 2;
if(argc < 2)
{
std::cout << "Usage: ./build/cl_events [input_image.ppm]\n\n";
std::cout << "No input_image provided, creating a dummy 640x480 image\n";
}
else
{
ppm.open(argv[1]);
}
TensorInfo dst_info(
src.info()->dimension(0) / scale_factor,
src.info()->dimension(1) / scale_factor,
Format::U8);
dst.allocator()->init(dst_info);
tmp_scale_median.allocator()->init(dst_info);
tmp_median_gauss.allocator()->init(dst_info);
src.allocator()->allocate();
dst.allocator()->allocate();
tmp_scale_median.allocator()->allocate();
tmp_median_gauss.allocator()->allocate();
if(ppm.is_open())
{
output_filename = std::string(argv[1]) + "_out.ppm";
}
OpenCL / Neon interoperability
You can mix OpenCL and Neon kernels and functions. However it is the user's responsibility to handle the mapping/unmapping of OpenCL objects, for example:
PPMLoader ppm;
if(argc < 2)
{
std::cout << "Usage: ./build/cl_convolution [input_image.ppm]\n\n";
std::cout << "No input_image provided, creating a dummy 640x480 image\n";
}
else
{
ppm.open(argv[1]);
}
TensorInfo scale_median_info(TensorInfo(
src.info()->dimension(0) / 2,
src.info()->dimension(1) / 2,
Format::U8));
scale_median.allocator()->init(scale_median_info);
median_gauss.allocator()->init(scale_median_info);
dst.allocator()->init(scale_median_info);
src.allocator()->allocate();
scale_median.allocator()->allocate();
median_gauss.allocator()->allocate();
dst.allocator()->allocate();
if(ppm.is_open())
{
const std::string output_filename = std::string(argv[1]) + "_out.ppm";
}
- See also
- main_neoncl_scale_median_gaussian
Algorithms
All computer vision algorithms in this library have been implemented following the OpenVX 1.1 specifications. Please refer to the Khronos documentation for more information.
Images, padding, border modes and tensors
Most kernels and functions in the library process images, however, in order to be future proof most of the kernels actually accept tensors. See below for more information about how they are related.
- Attention
- Each memory object can be written by only one kernel, however it can be read by several kernels. Writing to the same object from several kernels will result in undefined behavior. The kernel writing to an object must be configured before the kernel(s) reading from it.
Padding and border modes
Several algorithms require a neighborhood around the current pixel to compute it's value. This means the algorithm will not be able to process the borders of the image unless you give it more information about how those border pixels should be processed. The BorderMode enum is used for this purpose.
You have 3 types of BorderMode :
- BorderMode::UNDEFINED : Neighbor pixels outside of the image are treated as undefined. As a result all the pixels which are on the border will have a value which is undefined.
- BorderMode::REPLICATE : Neighbor pixels outside of the image are treated as having the same value as the closest valid pixel.
- BorderMode::CONSTANT : Neighbor pixels outside of the image are treated as having the same constant value. (The user can choose what this value should be).
Moreover both OpenCL and Neon use vector loads and stores instructions to access the data in buffers, so in order to avoid having special cases to handle for the borders all the images and tensors used in this library must be padded.
Padding
There are different ways padding can be calculated:
PPMLoader ppm;
if(argc < 2)
{
std::cout << "Usage: ./build/neon_convolution [input_image.ppm]\n\n";
std::cout << "No input_image provided, creating a dummy 640x480 image\n";
}
else
{
ppm.open(argv[1]);
}
tmp.allocator()->init(*
src.info());
dst.allocator()->init(*
src.info());
src.allocator()->allocate();
tmp.allocator()->allocate();
dst.allocator()->allocate();
if(ppm.is_open())
{
output_filename = std::string(argv[1]) + "_out.ppm";
}
- Note
- It's important to call allocate after the function is configured: if the image / tensor is already allocated then the function will shrink its execution window instead of increasing the padding. (See below for more details).
- Manual padding / no padding / auto padding: You can allocate your images / tensors up front (before configuring your functions). In that case the function will use whatever padding is available and will shrink its execution window if there isn't enough padding available (which translates into a smaller valid region for the output). See also Valid regions). If you don't want to manually set the padding but still want to allocate your objects upfront then you can use auto_padding. It guarantees that the allocation will have enough padding to run any of the provided functions.
src.info()->init_auto_padding(TensorShape(640u,480u),
Format::U8);
dst.info()->init(src.info()->tensor_shape(),
Format::U8, strides_in_bytes, offset_first_element_in_bytes, total_size_in_bytes);
src.allocator()->allocate();
dst.allocator()->allocate();
fill_image(src);
NEGaussian3x3 gauss;
gauss.run();
- Warning
- Some kernels need up to 3 neighbor values to calculate the value of a given pixel. Therefore, to be safe, we use a 4-pixel padding all around the image. In addition, some kernels read and write up to 32 pixels at the same time. To cover that case as well we add an extra 32 pixels of padding at the end of each row. As a result auto padded buffers waste a lot of memory and are less cache friendly. It is therefore recommended to use accurate padding or manual padding wherever possible.
Valid regions
Some kernels (like edge detectors for example) need to read values of neighboring pixels to calculate the value of a given pixel, it is therefore not possible to calculate the values of the pixels on the edges.
Another case is: if a kernel processes 8 pixels per iteration and the image's dimensions are not a multiple of 8 and not enough padding is available then the kernel will not be able to process the pixels near the right edge. As a result these pixels will be left undefined.
In order to know which pixels have been calculated, each kernel sets a valid region for each output image or tensor. See also TensorInfo::valid_region(), ValidRegion
Tensors
Tensors are multi-dimensional arrays with a maximum of Coordinates::num_max_dimensions dimensions.
Depending on the number of dimensions tensors can be interpreted as various objects. A scalar can be represented as a zero-dimensional tensor and a vector of numbers can be represented as an one-dimensional tensor. Further, an image is actually just a 2D tensor, a 3D tensor can be seen as an array of images and a 4D tensor as a 2D array of images, etc.
- Note
- Most algorithms process images (i.e a 2D slice of the tensor), therefore only padding along the X and Y axes is required (2D slices can be stored contiguously in memory).
Images and Tensors description conventions
Image objects are defined by a Format and dimensions expressed as [width, height, batch]
Tensors are defined by a DataType plus a number of channels (Always expected to be 1 for now) and their dimensions are expressed as [width, height, feature_maps, batch].
In other words, the lower three dimensions of a tensor specify a single input in [width, height, feature_maps], while any other specified dimension represents a batch in the appropriate dimension space. For example, a tensor with dimensions [128, 128, 64, 16] represents a 1D batch space with 16 batches of 128 elements in width and height and 64 feature maps each. Each kernel specifies the expected layout of each of its tensors in its documentation.
- Note
- Unless specified otherwise in the kernel's or function's documentation all tensors and images parameters passed must have identical dimensions.
-
Unless specified otherwise in the kernel's or function's documentation the number of channels for tensors is expected to be 1 (For images, the number of channels is inferred from the Format).
- Attention
- Regardless of the DataType used by a tensor the ITensor::buffer() method will always return a uint8_t pointer, and all the metadata in TensorInfo will be expressed in bytes. It is the user's responsibility to cast the pointer to the correct type.
For example, to read the element located at the coordinates (x,y) of a float tensor:
float value = *
reinterpret_cast<float*
>(
input.buffer() +
input.info()->offset_element_in_bytes(Coordinates(x,y)));
Working with Images and Tensors using iterators
The library provides some iterators to access objects' data. Iterators are created by associating a data object (An image or a tensor for example) with an iteration window.
Iteration windows are defined by an array of dimensions, each of which consists of a start, end and step.
The execute_window_loop function takes an execution window, a lambda function and one or more iterators. It will iterate through every element of the execution window and for each element it will update the iterators accordingly and call the lambda function.
Here are a couple of examples of how to use the iterators to fill / read tensors:
constexpr unsigned int width = 4;
constexpr unsigned int height = 3;
constexpr unsigned int batch = 2;
src_data = new float[width * height * batch];
dst_data = new float[width * height * batch];
for(
unsigned int b = 0;
b < batch;
b++)
{
for(unsigned int h = 0; h < height; h++)
{
for(
unsigned int w = 0;
w < width;
w++)
{
src_data[
b * (width * height) + h * width +
w] = static_cast<float>(100 *
b + 10 * h +
w);
}
}
}
const TensorShape
shape(width, height, batch);
softmax.configure(&
input, &output);
input.allocator()->allocate();
output.allocator()->allocate();
Window input_window;
input_window.use_tensor_dimensions(
input.info()->tensor_shape());
std::cout << " Dimensions of the input's iterator:\n";
std::cout << " X = [start=" << input_window.x().start() << ", end=" << input_window.x().end() << ", step=" << input_window.x().step() << "]\n";
std::cout << " Y = [start=" << input_window.y().start() << ", end=" << input_window.y().end() << ", step=" << input_window.y().step() << "]\n";
std::cout << " Z = [start=" << input_window.z().start() << ", end=" << input_window.z().end() << ", step=" << input_window.z().step() << "]\n";
Iterator input_it(&
input, input_window);
{
std::cout << "Setting item [" << id.x() << "," << id.y() << "," << id.z() << "]\n";
*reinterpret_cast<float *>(input_it.ptr()) = src_data[id.z() * (width * height) + id.y() * width + id.x()];
},
input_it);
Window output_window;
output_window.use_tensor_dimensions(output.info()->tensor_shape(),
Window::DimY);
std::cout << " Dimensions of the output's iterator:\n";
std::cout << " X = [start=" << output_window.x().start() << ", end=" << output_window.x().end() << ", step=" << output_window.x().step() << "]\n";
std::cout << " Y = [start=" << output_window.y().start() << ", end=" << output_window.y().end() << ", step=" << output_window.y().step() << "]\n";
std::cout << " Z = [start=" << output_window.z().start() << ", end=" << output_window.z().end() << ", step=" << output_window.z().step() << "]\n";
Iterator output_it(&output, output_window);
{
std::cout << "Copying one row starting from [" << id.x() << "," << id.y() << "," << id.z() << "]\n";
memcpy(dst_data + id.z() * (width * height) + id.y() * width, output_it.ptr(), width * sizeof(float));
},
output_it);
Sub-tensors
Sub-tensors are aliases to existing Tensors, as a result creating a sub-tensor does not result in any underlying memory allocation.
Sub-tensors can be used to access a sub-set of the parent tensor, something that can be useful in case different operations need to be performed on different parts of a tensor.
Moreover, sub-tensors can be used to perform zero copy tensor concatenation.
The API for creating a sub-tensor is the following:
SubTensor(ITensor *parent, const TensorShape &tensor_shape, const Coordinates &coords)
Where parent is the parent tensor which we want to create an alias for, tensor_shape is the shape of the sub-tensor and coords are the starting indexing coordinates of the sub-tensor within the parent tensor.
- Note
- Two sub-tensor concrete classes for different targets are currently supported : CLSubTensor and SubTensor
- Warning
- Limitation of the sub-tensor is that it cannot be extracted spatially, meaning sub-tensors should have the same width and height as the parent tensor. The main reasons for this is the fact that individual kernels might need to operate with a step size that is not a multiple of the sub-tensor spatial dimension. This could lead to elements being overwritten by different kernels operating on different sub-tensors of the same underlying tensor.
MemoryManager
IMemoryManager is a memory managing interface that can be used to reduce the memory requirements of a given pipeline by recycling temporary buffers.
MemoryGroup, MemoryPool and MemoryManager Components
MemoryGroup
IMemoryGroup defines the memory managing granularity.
MemoryGroup binds a number of objects to a bucket of memory requirements that need to be fulfilled in order for an operation or list of operations to be executed.
Requesting backing memory for a specific group can be done using IMemoryGroup::acquire and releasing the memory back using IMemoryGroup::release.
MemoryPool
IMemoryPool defines a pool of memory that can be used to provide backing memory to a memory group.
- Note
- BlobMemoryPool is currently implemented which models the memory requirements as a vector of distinct memory blobs.
MemoryManager Components
IMemoryManager consists of two components:
- ILifetimeManager that keeps track of the lifetime of the registered objects of the memory groups and given an IAllocator creates an appropriate memory pool that fulfils the memory requirements of all the registered memory groups.
- IPoolManager that safely manages the registered memory pools.
- Note
- BlobLifetimeManager is currently implemented which models the memory requirements as a vector of distinct memory blobs.
Working with the Memory Manager
Using a memory manager to reduce the memory requirements of a pipeline can be summed in the following steps:
Initially a memory manager must be set-up:
auto lifetime_mgr = std::make_shared<BlobLifetimeManager>();
auto pool_mgr = std::make_shared<PoolManager>();
auto mm = std::make_shared<MemoryManagerOnDemand>(lifetime_mgr, pool_mgr);
Once done, memory groups can be registered to use the memory manager:
MemoryGroup memory_group(mm);
- Note
- If a memory manager is not specified then all allocation will be immediate instead of deferred through the memory manager.
Next step is to set objects to be managed by the memory group. It is important though to note that the lifetime of an object is tracked from the MemoryGroup::manage() and the TensorAllocator::allocate calls. MemoryGroup::manage flags that the object will be needed starting now and when TensorAllocator::allocate is called it signals the end of the object lifetime.
Tensor tmp1, tmp2, tmp3;
memory_group.manage(&tmp1);
memory_group.manage(&tmp2);
operation1.configure(&tmp1, &tmp2);
tmp1.allocator()->allocate();
memory_group.manage(&tmp3);
operation2.configure(&tmp2, &tmp3);
tmp2.allocator()->allocate();
tmp3.allocator()->allocate();
- Warning
- The configuration step should be done sequentially by a single thread so that all the lifetimes are captured correclty.
When configuration of all the operations is finished then the memory manager have to be populated:
Finally, during execution of the pipeline the memory of the appropriate memory group should be requested before running:
memory_group.acquire();
operation1.run();
operation2.run();
memory_group.release();
- Note
- Execution of a pipeline can be done in a multi-threading environment as memory acquisition/release are thread safe.
-
If you are handling sensitive data and it's required to zero out the memory buffers before freeing, make sure to also zero out the intermediate buffers. You can access the buffers through the memory group's mappings.
Function support
Most of the library's function have been ported to use IMemoryManager for their internal temporary buffers.
If that is the case, a memory manager can be passed to them during construction to reuse memory among these functions.
auto lifetime_mgr = std::make_shared<BlobLifetimeManager>();
auto pool_mgr = std::make_shared<PoolManager>();
auto mm = std::make_shared<MemoryManagerOnDemand>(lifetime_mgr, pool_mgr);
CLConvolutionLayer conv1(mm), conv2(mm);
conv1.configure(...);
conv2.configure(...);
conv1.run();
conv2.run();
Import Memory Interface
The implemented TensorAllocator and CLTensorAllocator objects provide an interface capable of importing existing memory to a tensor as backing memory.
A simple Neon example can be the following:
void* external_ptr = ...;
Tensor tensor;
tensor.allocator()->init(tensor_info);
tensor.allocator()->import_memory(external_ptr);
It is important to note the following:
- Ownership of the backing memory is not transferred to the tensor itself.
- The tensor mustn't be memory managed.
- Padding requirements should be accounted by the client code. In other words, if padding is required by the tensor after the function configuration step, then the imported backing memory should account for it. Padding can be checked through the TensorInfo::padding() interface.
OpenCL Tuner
OpenCL kernels when dispatched to the GPU take two arguments:
- The Global Workgroup Size (GWS): That's the number of times to run an OpenCL kernel to process all the elements we want to process.
- The Local Workgroup Size (LWS): That's the number of elements we want to run in parallel on a GPU core at a given point in time.
The LWS can be required by an algorithm (For example if it contains memory barriers or uses local memory) but it can also be used for performance reasons to tweak the performance of a kernel: the execution time of the overall kernel might vary significantly depending on how the GWS is broken down.
However, there is no universal rule regarding which LWS is best for a given kernel, so instead we created the CLTuner.
When the CLTuner is enabled ( Target = 2 for the graph examples), the first time an OpenCL kernel is executed the Compute Library will try to run it with a variety of LWS values and will remember which one performed best for subsequent runs. At the end of the run the graph::Graph will try to save these tuning parameters to a file.
However this process takes quite a lot of time, which is why it cannot be enabled all the time. CLTuner supports three modes of tuning with different trade-offs between the time taken to tune and the kernel execution time achieved using the best LWS found. In the Exhaustive mode, it searches all the supported values of LWS. This mode takes the longest time to tune and is the most likely to find the optimal LWS. Normal mode searches a subset of LWS values to yield a good approximation of the optimal LWS. It takes less time to tune than Exhaustive mode. Rapid mode takes the shortest time to tune and finds an LWS value that is at least as good or better than the default LWS value. The mode affects only the search for the optimal LWS and has no effect when the LWS value is imported from a file.
But, when the CLTuner is disabled ( Target = 1 for the graph examples), the graph::Graph will try to reload the file containing the tuning parameters, then for each executed kernel the Compute Library will use the fine tuned LWS if it was present in the file or use a default LWS value if it's not.
Weights Manager
IWeightsManager is a weights managing interface that can be used to reduce the memory requirements of a given pipeline by reusing transformed weights across multiple function executions. IWeightsManager is responsible for managing weight tensors alongside with their transformations. ITransformWeights provides an interface for running the desired transform function. This interface is used by the weights manager.
Working with the Weights Manager
Following is a simple example that uses the weights manager:
Initially a weights manager must be set-up:
auto wm = std::make_shared<IWeightsManager>();
Once done, weights can be managed, configured and run:
wm->manage(weights);
wm->acquire(weights, &_reshape_weights_managed_function);
wm->run(weights, &_reshape_weights_managed_function);
Experimental Features
Run-time Context
Some of the Compute Library components are modelled as singletons thus posing limitations to supporting some use-cases and ensuring a more client-controlled API. Thus, we are introducing an aggregate service interface IRuntimeContext which will encapsulate the services that the singletons were providing and allow better control of these by the client code. Run-time context encapsulates a list of mechanisms, some of them are: scheduling, memory management, kernel caching and others. Consequently, this will allow finer control of these services among pipelines when Compute Library is integrated in higher level frameworks.
This feature introduces some changes to our API. All the kernels/functions will now accept a Runtime Context object which will allow the function to use the mentioned services.
Finally, we will try to adapt our code-base progressively to use the new mechanism but will continue supporting the legacy mechanism to allow a smooth transition. Changes will apply to all our three backends: Neon, OpenCL and OpenGL ES.
1.8.13