cuda-0.10.2.0: FFI binding to the CUDA interface for programming NVIDIA GPUs
Copyright[2018] Trevor L. McDonell
LicenseBSD
Safe HaskellNone
LanguageHaskell98

Foreign.CUDA.Driver.Graph.Exec

Contents

Description

Graph execution functions for the low-level driver interface

Requires CUDA-10

Synopsis

Documentation

newtype Executable Source #

Constructors

Executable 

Fields

Instances

Instances details
Eq Executable Source # 
Instance details

Defined in Foreign.CUDA.Driver.Graph.Base

Show Executable Source # 
Instance details

Defined in Foreign.CUDA.Driver.Graph.Base

Execution

launch :: Executable -> Stream -> IO () Source #

Execute a graph in the given stream. Only one instance may execute at a time; to execute a graph concurrently, it must be instantiated into multiple executables.

Requires CUDA-10.0

https://docs.nvidia.com/cuda/cuda-driver-api/group__CUDA__GRAPH.html#group__CUDA__GRAPH_1g6b2dceb3901e71a390d2bd8b0491e471

Since: 0.10.0.0

instantiate :: Graph -> IO Executable Source #

Instantiate the task graph description of a program into an executable graph.

Requires CUDA-10.0

https://docs.nvidia.com/cuda/cuda-driver-api/group__CUDA__GRAPH.html#group__CUDA__GRAPH_1g433ae118a751c9f2087f53d7add7bc2c

Since: 0.10.0.0

setKernel Source #

Arguments

:: Executable 
-> Node 
-> Fun 
-> (Int, Int, Int)

grid dimension

-> (Int, Int, Int)

thread block dimensions

-> Int

shared memory (bytes)

-> [FunParam] 
-> IO () 

Update the parameters for a kernel node in the given executable graph

Requires CUDA-10.1

https://docs.nvidia.com/cuda/cuda-driver-api/group__CUDA__GRAPH.html#group__CUDA__GRAPH_1gd84243569e4c3d6356b9f2eea20ed48c

Since: 0.10.1.0