CLFORK(3)             Standard Compute Layer (CL) Manual             CLFORK(3)



NAME
       clfork - Execute a CL kernel

SYNOPSIS
       #include <stdcl.h>

       cl_event  clfork(  CONTEXT*  cp,  unsigned  int  devnum, cl_kernel krn,
       clndrange_t* ndr, int flags);

       Link with -lstdcl.

DESCRIPTION
       clfork() is used to execute a CL kernel on a  specified  compute  layer
       (CL)  co-processor  device.   The  arguments for the kernel must be set
       prior to the call to clfork() using the  clarg_set*()  functions.   The
       kernel is executed over an index-space of work-items defined by ndr.

       The bahavior of clfork() can be controlled using the following flags:

       CL_EVENT_WAIT
              clfork() will block until the operation has completed.

       CL_EVENT_NOWAIT
              clfork()  will  return  immediately.  The programmer must ensure
              that the operation has completed using clwait() or clwaitev().

       CL_EVENT_RELEASE
              Used with CL_EVENT_WAIT to force  clfork()  to  release  the  CL
              event  generated by the operation.  If this flag is not used the
              programmer is responsible for releasing the returned event using
              clReleaseEvent().   This flag has no effect when CL_EVENT_NOWAIT
              is used.

RETURN VALUE
       On error clfork() will return (cl_event)(-1) and errno is set appropri-
       ately.

AUTHOR
       Written by David Richie.

REPORTING BUGS
       Report bugs to <support@browndeertechnology.com>

COPYRIGHT
       Copyright  (C) 2009 Brown Deer Technology, LLC.  Licensed under the GNU
       Lesser General Public License version 3 (LGPLv3).  There is NO WARRANTY
       to the extent permitted by law.

SEE ALSO
       clarg_set(3),     clndrange_init(3),    clndrange_set(3),    clwait(3),
       clwaitev(3) stdcl(3)



libstdcl-0.7                      2009-11-10                         CLFORK(3)