futhark-0.19.7: An optimising compiler for a functional, array-oriented language.
Safe HaskellNone
LanguageHaskell2010

Futhark.CodeGen.ImpGen.GPU

Description

Compile a GPUMem program to imperative code with kernels. This is mostly (but not entirely) the same process no matter if we are targeting OpenCL or CUDA. The important distinctions (the host level code) are introduced later.

Synopsis

Documentation

compileProgOpenCL :: MonadFreshNames m => Prog GPUMem -> m (Warnings, Program) Source #

Compile a GPUMem program to low-level parallel code, with either CUDA or OpenCL characteristics.

compileProgCUDA :: MonadFreshNames m => Prog GPUMem -> m (Warnings, Program) Source #

Compile a GPUMem program to low-level parallel code, with either CUDA or OpenCL characteristics.

data Warnings Source #

The warnings produced by the compiler. The Show instance produces a human-readable description.

Instances

Instances details
Semigroup Warnings Source # 
Instance details

Defined in Language.Futhark.Warnings

Monoid Warnings Source # 
Instance details

Defined in Language.Futhark.Warnings

Pretty Warnings Source # 
Instance details

Defined in Language.Futhark.Warnings

Methods

ppr :: Warnings -> Doc #

pprPrec :: Int -> Warnings -> Doc #

pprList :: [Warnings] -> Doc #