Futhark.CodeGen.ImpGen.Kernels.SegScan.SinglePass

Code generation for segmented and non-segmented scans. Uses a fast single-pass algorithm, but which only works on NVIDIA GPUs and with some constraints on the operator. We use this when we can.

Documentation

compileSegScan :: Pattern KernelsMem -> SegLevel -> SegSpace -> SegBinOp KernelsMem -> KernelBody KernelsMem -> CallKernelGen ()

Compile SegScan instance to host-level code with calls to a single-pass kernel.