starpu-max/doc/doxygen/chapters/api/initialization.doxy

/*
 * This file is part of the StarPU Handbook.
 * Copyright (C) 2009--2011  Universit@'e de Bordeaux
 * Copyright (C) 2010, 2011, 2012, 2013, 2014, 2016, 2017  CNRS
 * Copyright (C) 2011, 2012, 2017  INRIA
 * See the file version.doxy for copying conditions.
 */

/*! \defgroup API_Initialization_and_Termination Initialization and Termination

\struct starpu_conf
\ingroup API_Initialization_and_Termination
This structure is passed to the starpu_init() function in order to
configure StarPU. It has to be initialized with starpu_conf_init().
When the default value is used, StarPU automatically selects the
number of processing units and takes the default scheduling policy.
The environment variables overwrite the equivalent parameters.
\var int starpu_conf::magic
\private
    Will be initialized by starpu_conf_init(). Should not be set by
    hand.

\var const char*starpu_conf::sched_policy_name
    Name of the scheduling policy. This can also be specified with the
    environment variable \ref STARPU_SCHED. (default = <c>NULL</c>).

\var struct starpu_sched_policy *starpu_conf::sched_policy
    Definition of the scheduling policy. This field is ignored if
    starpu_conf::sched_policy_name is set. (default = <c>NULL</c>)

\var void (*starpu_conf::sched_policy_init)(unsigned)
    todo

\var int starpu_conf::ncpus
    Number of CPU cores that StarPU can use. This can also be
    specified with the environment variable \ref STARPU_NCPU .
    (default = -1)
\var int starpu_conf::ncuda
    Number of CUDA devices that StarPU can use. This can also be
    specified with the environment variable \ref STARPU_NCUDA.
    (default = -1)
\var int starpu_conf::nopencl
    Number of OpenCL devices that StarPU can use. This can also be
    specified with the environment variable \ref STARPU_NOPENCL.
    (default = -1)
\var int starpu_conf::nmic
    Number of MIC devices that StarPU can use. This can also be
    specified with the environment variable \ref STARPU_NMIC.
    (default = -1)
\var int starpu_conf::nscc
    Number of SCC devices that StarPU can use. This can also be
    specified with the environment variable \ref STARPU_NSCC.
    (default = -1)
\var int starpu_conf::nmpi_ms
    Number of MPI Master Slave devices that StarPU can use. This can
    also be specified with the environment variable \ref
    STARPU_NMPI_MS. (default = -1)

\var unsigned starpu_conf::use_explicit_workers_bindid
    If this flag is set, the starpu_conf::workers_bindid array
    indicates where the different workers are bound, otherwise StarPU
    automatically selects where to bind the different workers. This
    can also be specified with the environment variable \ref
    STARPU_WORKERS_CPUID. (default = 0)
\var unsigned starpu_conf::workers_bindid[STARPU_NMAXWORKERS]
    If the starpu_conf::use_explicit_workers_bindid flag is set, this
    array indicates where to bind the different workers. The i-th
    entry of the starpu_conf::workers_bindid indicates the logical
    identifier of the processor which should execute the i-th worker.
    Note that the logical ordering of the CPUs is either determined by
    the OS, or provided by the hwloc library in case it is available.
\var unsigned starpu_conf::use_explicit_workers_cuda_gpuid
    If this flag is set, the CUDA workers will be attached to the CUDA
    devices specified in the starpu_conf::workers_cuda_gpuid array.
    Otherwise, StarPU affects the CUDA devices in a round-robin
    fashion. This can also be specified with the environment variable
    \ref STARPU_WORKERS_CUDAID. (default = 0)
\var unsigned starpu_conf::workers_cuda_gpuid[STARPU_NMAXWORKERS]
    If the starpu_conf::use_explicit_workers_cuda_gpuid flag is set,
    this array contains the logical identifiers of the CUDA devices
    (as used by \c cudaGetDevice()).
\var unsigned starpu_conf::use_explicit_workers_opencl_gpuid
    If this flag is set, the OpenCL workers will be attached to the
    OpenCL devices specified in the starpu_conf::workers_opencl_gpuid
    array. Otherwise, StarPU affects the OpenCL devices in a
    round-robin fashion. This can also be specified with the
    environment variable \ref STARPU_WORKERS_OPENCLID. (default = 0)
\var unsigned starpu_conf::workers_opencl_gpuid[STARPU_NMAXWORKERS]
    If the starpu_conf::use_explicit_workers_opencl_gpuid flag is set,
    this array contains the logical identifiers of the OpenCL devices
    to be used.
\var unsigned starpu_conf::use_explicit_workers_mic_deviceid
    If this flag is set, the MIC workers will be attached to the MIC
    devices specified in the array starpu_conf::workers_mic_deviceid.
    Otherwise, StarPU affects the MIC devices in a round-robin
    fashion. This can also be specified with the environment variable
    \ref STARPU_WORKERS_MICID. (default = 0)
\var unsigned starpu_conf::workers_mic_deviceid[STARPU_NMAXWORKERS]
    If the flag starpu_conf::use_explicit_workers_mic_deviceid is set,
    the array contains the logical identifiers of the MIC devices to
    be used.
\var unsigned starpu_conf::use_explicit_workers_scc_deviceid
    If this flag is set, the SCC workers will be attached to the SCC
    devices specified in the array starpu_conf::workers_scc_deviceid.
    (default = 0)
\var unsigned starpu_conf::workers_scc_deviceid[STARPU_NMAXWORKERS]
    If the flag starpu_conf::use_explicit_workers_scc_deviceid is set,
    the array contains the logical identifiers of the SCC devices to
    be used. Otherwise, StarPU affects the SCC devices in a
    round-robin fashion. This can also be specified with the
    environment variable \ref STARPU_WORKERS_SCCID.
\var unsigned starpu_conf::use_explicit_workers_mpi_ms_deviceid
    If this flag is set, the MPI Master Slave workers will be attached
    to the MPI Master Slave devices specified in the array
    starpu_conf::workers_mpi_ms_deviceid. Otherwise, StarPU affects
    the MPI Master Slave devices in a round-robin fashion. (default =
    0)
\var unsigned starpu_conf::workers_mpi_ms_deviceid[STARPU_NMAXWORKERS]
    If the flag starpu_conf::use_explicit_workers_mpi_ms_deviceid is
    set, the array contains the logical identifiers of the MPI Master
    Slave devices to be used.

\var int starpu_conf::bus_calibrate
    If this flag is set, StarPU will recalibrate the bus.  If this
    value is equal to -1, the default value is used. This can
    also be specified with the environment variable \ref
    STARPU_BUS_CALIBRATE. (default = 0)
\var int starpu_conf::calibrate
    If this flag is set, StarPU will calibrate the performance models
    when executing tasks. If this value is equal to -1, the
    default value is used. If the value is equal to 1, it will
    force continuing calibration. If the value is equal to 2,
    the existing performance models will be overwritten. This can also
    be specified with the environment variable \ref STARPU_CALIBRATE.
    (default = 0)
\var int starpu_conf::single_combined_worker
    By default, StarPU executes parallel tasks concurrently. Some
    parallel libraries (e.g. most OpenMP implementations) however do
    not support concurrent calls to parallel code. In such case,
    setting this flag makes StarPU only start one parallel task at a
    time (but other CPU and GPU tasks are not affected and can be run
    concurrently). The parallel task scheduler will however still try
    varying combined worker sizes to look for the most efficient ones.
    This can also be specified with the environment variable \ref
    STARPU_SINGLE_COMBINED_WORKER. (default = 0)

\var char *starpu_conf::mic_sink_program_path
    Path to the kernel to execute on the MIC device, compiled for MIC
    architecture. When set to <c>NULL</c>, StarPU automatically looks
    next to the host program location. (default = <c>NULL</c>)

\var int starpu_conf::disable_asynchronous_copy
    This flag should be set to 1 to disable asynchronous copies
    between CPUs and all accelerators. This can also be specified with
    the environment variable \ref STARPU_DISABLE_ASYNCHRONOUS_COPY.
    The AMD implementation of OpenCL is known to fail when copying
    data asynchronously. When using this implementation, it is
    therefore necessary to disable asynchronous data transfers. This
    can also be specified at compilation time by giving to the
    configure script the option
    \ref disable-asynchronous-copy "--disable-asynchronous-copy".
    (default = 0)
\var int starpu_conf::disable_asynchronous_cuda_copy
    This flag should be set to 1 to disable asynchronous copies
    between CPUs and CUDA accelerators. This can also be specified
    with the environment variable \ref
    STARPU_DISABLE_ASYNCHRONOUS_CUDA_COPY.
    This can also be specified at compilation time by giving to the
    configure script the option
    \ref disable-asynchronous-cuda-copy "--disable-asynchronous-cuda-copy".
    (default = 0)
\var int starpu_conf::disable_asynchronous_opencl_copy
    This flag should be set to 1 to disable asynchronous copies
    between CPUs and OpenCL accelerators. This can also be specified
    with the environment variable \ref
    STARPU_DISABLE_ASYNCHRONOUS_OPENCL_COPY. The AMD implementation of
    OpenCL is known to fail when copying data asynchronously. When
    using this implementation, it is therefore necessary to disable
    asynchronous data transfers. This can also be specified at
    compilation time by giving to the configure script the
    option
    \ref disable-asynchronous-opencl-copy "--disable-asynchronous-opencl-copy".
    (default = 0)
\var int starpu_conf::disable_asynchronous_mic_copy
    This flag should be set to 1 to disable asynchronous copies
    between CPUs and MIC accelerators. This can also be specified with
    the environment variable \ref
    STARPU_DISABLE_ASYNCHRONOUS_MIC_COPY. This can also be specified
    at compilation time by giving to the configure script the option
    \ref disable-asynchronous-mic-copy "--disable-asynchronous-mic-copy".
    (default = 0).
\var int starpu_conf::disable_asynchronous_mpi_ms_copy
    This flag should be set to 1 to disable asynchronous copies
    between CPUs and MPI Master Slave devices. This can also be
    specified with the environment variable \ref
    STARPU_DISABLE_ASYNCHRONOUS_MPI_MS_COPY. This can also be
    specified at compilation time by giving to the configure script
    the option
    \ref disable-asynchronous-mpi-master-slave-copy "--disable-asynchronous-mpi-master-slave-copy".
    (default = 0).

\var unsigned *starpu_conf::cuda_opengl_interoperability
    Enable CUDA/OpenGL interoperation on these CUDA devices. This can
    be set to an array of CUDA device identifiers for which
    \c cudaGLSetGLDevice() should be called instead of
    \c cudaSetDevice(). Its size is specified by the
    starpu_conf::n_cuda_opengl_interoperability field below
    (default = <c>NULL</c>)
\var unsigned starpu_conf::n_cuda_opengl_interoperability
    todo

\var struct starpu_driver *starpu_conf::not_launched_drivers
    Array of drivers that should not be launched by StarPU. The
    application will run in one of its own threads. (default =
    <c>NULL</c>)
\var unsigned starpu_conf::n_not_launched_drivers
    The number of StarPU drivers that should not be launched by
    StarPU. (default = 0)

\var starpu_conf::trace_buffer_size
    Specify the buffer size used for FxT tracing. Starting from FxT
    version 0.2.12, the buffer will automatically be flushed when it
    fills in, but it may still be interesting to specify a bigger
    value to avoid any flushing (which would disturb the trace).

\var starpu_conf::global_sched_ctx_min_priority
    todo
\var starpu_conf::global_sched_ctx_max_priority
    todo

\fn int starpu_init(struct starpu_conf *conf)
\ingroup API_Initialization_and_Termination
This is StarPU initialization method, which must be called prior to
any other StarPU call. It is possible to specify StarPU’s
configuration (e.g. scheduling policy, number of cores, ...) by
passing a non-<c>NULL</c> \p conf. Default configuration is used if \p
conf is <c>NULL</c>. Upon successful completion, this function
returns 0. Otherwise, <c>-ENODEV</c> indicates that no worker was
available (and thus StarPU was not initialized).

\fn int starpu_initialize(struct starpu_conf *user_conf, int *argc, char ***argv)
\ingroup API_Initialization_and_Termination
This is the same as starpu_init(), but also takes the \p argc and \p
argv as defined by the application. This is needed for SCC execution
to initialize the communication library.
Do not call starpu_init() and starpu_initialize() in the
same program.

\fn int starpu_conf_init(struct starpu_conf *conf)
\ingroup API_Initialization_and_Termination
Initialize the \p conf structure with the default values. In case some
configuration parameters are already
specified through environment variables, starpu_conf_init() initializes
the fields of \p conf according to the environment variables.
For instance if \ref STARPU_CALIBRATE is set, its value is put in the
field starpu_conf::calibrate of \p conf. Upon successful
completion, this function returns 0. Otherwise, <c>-EINVAL</c> indicates that
the argument was <c>NULL</c>.

\fn void starpu_shutdown(void)
\ingroup API_Initialization_and_Termination
This is StarPU termination method. It must be called at the end of the
application: statistics and other post-mortem debugging information
are not guaranteed to be available until this method has been called.

\fn void starpu_pause(void)
\ingroup API_Initialization_and_Termination
Suspend the processing of new tasks by
workers. It can be used in a program where StarPU is used during only
a part of the execution. Without this call, the workers continue to
poll for new tasks in a tight loop, wasting CPU time. The symmetric
call to starpu_resume() should be used to unfreeze the workers.

\fn void starpu_resume(void)
\ingroup API_Initialization_and_Termination
This is the symmetrical call to starpu_pause(), used to resume
the workers polling for new tasks.

\fn int starpu_asynchronous_copy_disabled(void)
\ingroup API_Initialization_and_Termination
Return 1 if asynchronous data transfers between CPU and accelerators
are disabled.

\fn int starpu_asynchronous_cuda_copy_disabled(void)
\ingroup API_Initialization_and_Termination
Return 1 if asynchronous data transfers between CPU and CUDA
accelerators are disabled.

\fn int starpu_asynchronous_opencl_copy_disabled(void)
\ingroup API_Initialization_and_Termination
Return 1 if asynchronous data transfers between CPU and OpenCL
accelerators are disabled.

\fn int starpu_asynchronous_mic_copy_disabled(void)
\ingroup API_Initialization_and_Termination
Return 1 if asynchronous data transfers between CPU and MIC
devices are disabled.

\fn int starpu_asynchronous_mpi_ms_copy_disabled(void)
\ingroup API_Initialization_and_Termination
Return 1 if asynchronous data transfers between CPU and MPI Slave
devices are disabled.

\fn void starpu_topology_print(FILE *f)
\ingroup API_Initialization_and_Termination
Print a description of the topology on \p f.

*/