@c -*-texinfo-*-

@c This file is part of the StarPU Handbook.
@c Copyright (C) 2011 Institut National de Recherche en Informatique et Automatique
@c See the file starpu.texi for copying conditions.

@node C Extensions
@chapter C Extensions

@cindex C extensions
@cindex GCC plug-in

When configured with @code{--enable-gcc-extensions}, StarPU builds a
plug-in for the GNU Compiler Collection (GCC), which defines extensions
to languages of the C family (C, C++, Objective-C) that make it easier
to write StarPU code@footnote{This feature is only available for GCC 4.5
and later.}.  Those extensions include syntactic sugar for defining
tasks and their implementations, invoking a task, and manipulating data
buffers.  Use of these extensions can be made conditional on the
availability of the plug-in, leading to valid C sequential code when the
plug-in is not used (@pxref{Conditional Extensions}).

This section does not require detailed knowledge of the StarPU library.

Note: as of StarPU @value{VERSION}, this is still an area under
development and subject to change.

@menu
* Defining Tasks::              Defining StarPU tasks
* Registered Data Buffers::     Manipulating data buffers
* Conditional Extensions::      Using C extensions only when available
@end menu

@node Defining Tasks
@section Defining Tasks

@cindex task
@cindex task implementation

The StarPU GCC plug-in views @dfn{tasks} as ``extended'' C functions:

@enumerate
@item
tasks may have several implementations---e.g., one for CPUs, one written
in OpenCL, one written in CUDA;
@item
tasks may have several implementations of the same target---e.g.,
several CPU implementations;
@item
when a task is invoked, it may run in parallel, and StarPU is free to
choose any of its implementations.
@end enumerate

Tasks and their implementations must be @emph{declared}.  These
declarations are annotated with @dfn{attributes} (@pxref{Attribute
Syntax, attributes in GNU C,, gcc, Using the GNU Compiler Collection
(GCC)}): the declaration of a task is a regular C function declaration
with an additional @code{task} attribute, and task implementations are
declared with a @code{task_implementation} attribute.

The following function attributes are provided:

@table @code

@item task
@cindex @code{task} attribute
Declare the given function as a StarPU task.  Its return type must be
@code{void}, and it must not be defined---instead, a definition will
automatically be provided by the compiler.

Under the hood, declaring a task leads to the declaration of the
corresponding @code{codelet} (@pxref{Codelet and Tasks}).  If one or
more task implementations are declared in the same compilation unit,
then the codelet and the function itself are also defined; they inherit
the scope of the task.

Scalar arguments to the task are passed by value and copied to the
target device if need be---technically, they are passed as the
@code{cl_arg} buffer (@pxref{Codelets and Tasks, @code{cl_arg}}).

@cindex @code{output} type attribute
Pointer arguments are assumed to be registered data buffers---the
@code{buffers} argument of a task (@pxref{Codelets and Tasks,
@code{buffers}}); @code{const}-qualified pointer arguments are viewed as
read-only buffers (@code{STARPU_R}), and non-@code{const}-qualified
buffers are assumed to be used read-write (@code{STARPU_RW}).  In
addition, the @code{output} type attribute can be as a type qualifier
for output pointer or array parameters (@code{STARPU_W}).

@item task_implementation (@var{target}, @var{task})
@cindex @code{task_implementation} attribute
Declare the given function as an implementation of @var{task} to run on
@var{target}.  @var{target} must be a string, currently one of
@code{"cpu"} or @code{"cuda"}.
@c FIXME: Update when OpenCL support is ready.

@end table

Here is an example:

@example
#define __output  __attribute__ ((output))

static void matmul (const float *A, const float *B,
                    __output float *C,
                    size_t nx, size_t ny, size_t nz)
  __attribute__ ((task));

static void matmul_cpu (const float *A, const float *B,
                        __output float *C,
                        size_t nx, size_t ny, size_t nz)
  __attribute__ ((task_implementation ("cpu", matmul)));


static void
matmul_cpu (const float *A, const float *B, __output float *C,
            size_t nx, size_t ny, size_t nz)
@{
  size_t i, j, k;

  for (j = 0; j < ny; j++)
    for (i = 0; i < nx; i++)
      @{
        for (k = 0; k < nz; k++)
          C[j * nx + i] += A[j * nz + k] * B[k * nx + i];
      @}
@}
@end example

@noindent
A @code{matmult} task is defined; it has only one implementation,
@code{matmult_cpu}, which runs on the CPU.  Variables @var{A} and
@var{B} are input buffers, whereas @var{C} is considered an input/output
buffer.

CUDA and OpenCL implementations can be declared in a similar way:

@example
static void matmul_cuda (const float *A, const float *B, float *C,
                         size_t nx, size_t ny, size_t nz)
  __attribute__ ((task_implementation ("cuda", matmul)));

static void matmul_opencl (const float *A, const float *B, float *C,
                           size_t nx, size_t ny, size_t nz)
  __attribute__ ((task_implementation ("opencl", matmul)));
@end example

@noindent
The CUDA and OpenCL implementations typically either invoke a kernel
written in CUDA or OpenCL (for similar code, @pxref{CUDA Kernel}, and
@pxref{OpenCL Kernel}), or call a library function that uses CUDA or
OpenCL under the hood, such as CUBLAS functions:

@example
static void
matmul_cuda (const float *A, const float *B, float *C,
             size_t nx, size_t ny, size_t nz)
@{
  cublasSgemm ('n', 'n', nx, ny, nz,
               1.0f, A, 0, B, 0,
               0.0f, C, 0);
  cudaStreamSynchronize (starpu_cuda_get_local_stream ());
@}
@end example

A task can be invoked like a regular C function:

@example
matmul (&A[i * zdim * bydim + k * bzdim * bydim],
        &B[k * xdim * bzdim + j * bxdim * bzdim],
        &C[i * xdim * bydim + j * bxdim * bydim],
        bxdim, bydim, bzdim);
@end example

@noindent
This leads to an @dfn{asynchronous invocation}, whereby @code{matmult}'s
implementation may run in parallel with the continuation of the caller.

The next section describes how memory buffers must be handled in
StarPU-GCC code.


@node Registered Data Buffers
@section Registered Data Buffers

Data buffers such as matrices and vectors that are to be passed to tasks
must be @dfn{registered}.  Registration allows StarPU to handle data
transfers among devices---e.g., transferring an input buffer from the
CPU's main memory to a task scheduled to run a GPU (@pxref{StarPU Data
Management Library}).

The following pragmas are provided:

@table @code

@item #pragma starpu register @var{ptr} [@var{size}]
Register @var{ptr} as a @var{size}-element buffer.

@item #pragma starpu unregister @var{ptr}
@item #pragma starpu acquire @var{ptr}

@end table

FIXME: finish


@node Conditional Extensions
@section Using C Extensions Conditionally

The C extensions described in this chapter are only available when GCC
and its StarPU plug-in are in use.  Yet, it is possible to make use of
these extensions when they are available---leading to hybrid CPU/GPU
code---and discard them when they are not available---leading to valid
sequential code.

To that end, the GCC plug-in defines a C preprocessor macro when it is
being used:

@defmac STARPU_GCC_PLUGIN
Defined for code being compiled with the StarPU GCC plug-in.  When
defined, this macro expands to an integer denoting the version of the
supported C extensions.
@end defmac

The code below illustrates how to define a task and its implementations
in a way that allows it to be compiled without the GCC plug-in:

@example
/* The macros below abstract over the attributes specific to
   StarPU-GCC and the name of the CPU implementation.  */
#ifdef STARPU_GCC_PLUGIN
# define __task  __attribute__ ((task))
# define CPU_TASK_IMPL(task)  task ## _cpu
#else
# define __task
# define CPU_TASK_IMPL(task)  task
#endif

#include <stdlib.h>

static void matmul (const float *A, const float *B, float *C,
                    size_t nx, size_t ny, size_t nz) __task;

#ifdef STARPU_GCC_PLUGIN

static void matmul_cpu (const float *A, const float *B, float *C,
                        size_t nx, size_t ny, size_t nz)
  __attribute__ ((task_implementation ("cpu", matmul)));

#endif


static void
CPU_TASK_IMPL (matmul) (const float *A, const float *B, float *C,
                        size_t nx, size_t ny, size_t nz)
@{
  /* Code of the CPU kernel here...  */
@}

int
main (int argc, char *argv[])
@{
  /* The pragmas below are simply ignored when StarPU-GCC
     is not used.  */
#pragma starpu initialize

  float A[123][42][7], B[123][42][7], C[123][42][7];

#pragma starpu register A
#pragma starpu register B
#pragma starpu register C

  /* When StarPU-GCC is used, the call below is asynchronous;
     otherwise, it is synchronous.  */
  matmul (A, B, C, 123, 42, 7);

#pragma starpu wait
#pragma starpu shutdown

  return EXIT_SUCCESS;
@}
@end example

Note that attributes such as @code{task} are simply ignored by GCC when
the StarPU plug-in is not loaded, so the @code{__task} macro could be
omitted altogether.  However, @command{gcc -Wall} emits a warning for
unknown attributes, which can be inconvenient, and other compilers may
be unable to parse the attribute syntax.  Thus, using macros such as
@code{__task} above is recommended.

@c Local Variables:
@c TeX-master: "../starpu.texi"
@c ispell-local-dictionary: "american"
@c End: