forked from OSchip/llvm-project
24 lines
1016 B
Plaintext
24 lines
1016 B
Plaintext
// RUN: %clang_cc1 -triple nvptx64-unknown-unknown -target-cpu sm_60 \
|
|
// RUN: -fcuda-is-device -S -emit-llvm -o - -x cuda %s \
|
|
// RUN: | FileCheck -check-prefix=CHECK %s
|
|
//
|
|
// RUN: %clang_cc1 -triple nvptx-unknown-unknown -target-cpu sm_50 \
|
|
// RUN: -fcuda-is-device -S -o /dev/null -x cuda -verify %s
|
|
|
|
#define __device__ __attribute__((device))
|
|
#define __global__ __attribute__((global))
|
|
#define __shared__ __attribute__((shared))
|
|
#define __constant__ __attribute__((constant))
|
|
|
|
// We have to keep all builtins that depend on particular target feature in the
|
|
// same function, because the codegen will stop after the very first function
|
|
// that encounters an error, so -verify will not be able to find errors in
|
|
// subsequent functions.
|
|
|
|
// CHECK-LABEL: test_fn
|
|
__device__ void test_fn(double d, double* double_ptr) {
|
|
// CHECK: call double @llvm.nvvm.atomic.load.add.f64.p0f64
|
|
// expected-error@+1 {{'__nvvm_atom_add_gen_d' needs target feature sm_60}}
|
|
__nvvm_atom_add_gen_d(double_ptr, d);
|
|
}
|