docc 0.8.3 (0d44999b)
C++ API Reference
Loading...
Searching...
No Matches
sdfg::cuda::ScheduleType_CUDA Class Reference

CUDA schedule type inheriting shared GPU functionality Provides CUDA-specific value() and default block size (32 for warp size). More...

#include <sdfg/targets/cuda/cuda.h>

Inheritance diagram for sdfg::cuda::ScheduleType_CUDA:
[legend]

Static Public Member Functions

static const std::string value ()
static symbolic::Integer default_block_size_x ()
Static Public Member Functions inherited from sdfg::gpu::ScheduleType_GPU_Base< ScheduleType_CUDA >
static void dimension (structured_control_flow::ScheduleType &schedule, const GPUDimension &dimension)
 Set the GPU dimension for a schedule.
static void block_size (structured_control_flow::ScheduleType &schedule, const symbolic::Expression block_size)
 Set the block size for a schedule.
static bool nested_sync (const structured_control_flow::ScheduleType &schedule)
 Check if nested synchronization is enabled.
static structured_control_flow::ScheduleType create ()
 Create a new GPU schedule type.

Detailed Description

CUDA schedule type inheriting shared GPU functionality Provides CUDA-specific value() and default block size (32 for warp size).

Deprecated
This class is deprecated and will be removed in future versions. Use the new GPU schedule type classes instead.

Member Function Documentation

◆ default_block_size_x()

symbolic::Integer sdfg::cuda::ScheduleType_CUDA::default_block_size_x ( )
inlinestatic

◆ value()

const std::string sdfg::cuda::ScheduleType_CUDA::value ( )
inlinestatic

The documentation for this class was generated from the following file:
  • opt/include/sdfg/targets/cuda/cuda.h