Revision 2b3aaa8d6a5bb54a3da4e32395bb80ca09b4e4b3 authored by xndcn on 08 January 2021, 19:46:56 UTC, committed by GitHub on 08 January 2021, 19:46:56 UTC
* Add max threads checking for Metal

Originally, this checking will be asserted by Metal API Validation
in Xcode, otherwise the program will crash or output wrong results.

* Disable the max threads checking for Metal in non-debug runtime

* Disable error/metal_threads_too_large test for non-OSX target
1 parent 081f472
Raw File
SelectGPUAPI.h
#ifndef HALIDE_INTERNAL_SELECT_GPU_API_H
#define HALIDE_INTERNAL_SELECT_GPU_API_H

#include "Expr.h"
#include "Target.h"

/** \file
 * Defines a lowering pass that selects which GPU api to use for each
 * gpu for loop
 */

namespace Halide {
namespace Internal {

/** Replace for loops with GPU_Default device_api with an actual
 * device API depending on what's enabled in the target. Choose the
 * first of the following: opencl, cuda, openglcompute, opengl */
Stmt select_gpu_api(const Stmt &s, Target t);

}  // namespace Internal
}  // namespace Halide

#endif
back to top