Content - 4486da4ebc9698019ac257a9796b1f114d1b4121 - 2dbeb1f/src/CodeGen_PTX_Dev.h

visit type:

Tip revision: ded839230e274e8d017916338e569272d6c8c6e2 authored by Aelphy on 01 February 2024, 21:58:14 UTC
[xtensa] replaced convert int16->int32_x2->int16 to two interleavs for better efficiency

Tip revision: ded8392

CodeGen_PTX_Dev.h

#ifndef HALIDE_CODEGEN_PTX_DEV_H
#define HALIDE_CODEGEN_PTX_DEV_H

/** \file
 * Defines the code-generator for producing CUDA host code
 */

#include <memory>

namespace Halide {

struct Target;

namespace Internal {

struct CodeGen_GPU_Dev;

std::unique_ptr<CodeGen_GPU_Dev> new_CodeGen_PTX_Dev(const Target &target);

}  // namespace Internal
}  // namespace Halide

#endif

Browse the archive

https://github.com/halide/Halide