338dee0742
With this PR, if we have customized implementation for scalar or vector length = 2, we don't need to write new macros, e.g. https://github.com/intel/llvm/blob/fb18321705f6/libclc/clc/include/clc/clcmacro.h#L15 Undef __HALF_ONLY, __FLOAT_ONLY and __DOUBLE_ONLY at the end of clc/include/clc/math/gentype.inc llvm-diff shows no change to nvptx64--nvidiacl.bc and amdgcn--amdhsa.bc
35 lines
1.1 KiB
Common Lisp
35 lines
1.1 KiB
Common Lisp
//===----------------------------------------------------------------------===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include <clc/clcmacro.h>
|
|
#include <clc/internal/clc.h>
|
|
#include <clc/math/clc_ldexp.h>
|
|
|
|
#define FUNCTION __clc_ldexp
|
|
#define __CLC_ARG2_TYPE int
|
|
#define __CLC_MIN_VECSIZE 1
|
|
|
|
#ifdef __HAS_LDEXPF__
|
|
// This defines all the ldexp(floatN, intN) variants.
|
|
#define __FLOAT_ONLY
|
|
#define __IMPL_FUNCTION __builtin_amdgcn_ldexpf
|
|
#define __CLC_BODY <clc/shared/binary_def_scalarize.inc>
|
|
#include <clc/math/gentype.inc>
|
|
#undef __IMPL_FUNCTION
|
|
#endif
|
|
|
|
#ifdef cl_khr_fp64
|
|
#pragma OPENCL EXTENSION cl_khr_fp64 : enable
|
|
// This defines all the ldexp(doubleN, intN) variants.
|
|
#define __DOUBLE_ONLY
|
|
#define __IMPL_FUNCTION __builtin_amdgcn_ldexp
|
|
#define __CLC_BODY <clc/shared/binary_def_scalarize.inc>
|
|
#include <clc/math/gentype.inc>
|
|
#undef __IMPL_FUNCTION
|
|
#endif
|