File: flush-denormals.cu

package info (click to toggle)
llvm-toolchain-6.0 1%3A6.0.1-10
  • links: PTS, VCS
  • area: main
  • in suites: buster
  • size: 598,080 kB
  • sloc: cpp: 3,046,253; ansic: 595,057; asm: 271,965; python: 128,926; objc: 106,554; sh: 21,906; lisp: 10,191; pascal: 6,094; ml: 5,544; perl: 5,265; makefile: 2,227; cs: 2,027; xml: 686; php: 212; csh: 117
file content (25 lines) | stat: -rw-r--r-- 1,077 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
// RUN: %clang_cc1 -fcuda-is-device \
// RUN:   -triple nvptx-nvidia-cuda -emit-llvm -o - %s | \
// RUN:   FileCheck %s -check-prefix CHECK -check-prefix NOFTZ
// RUN: %clang_cc1 -fcuda-is-device -fcuda-flush-denormals-to-zero \
// RUN:   -triple nvptx-nvidia-cuda -emit-llvm -o - %s | \
// RUN:   FileCheck %s -check-prefix CHECK -check-prefix FTZ

#include "Inputs/cuda.h"

// Checks that device function calls get emitted with the "ntpvx-f32ftz"
// attribute set to "true" when we compile CUDA device code with
// -fcuda-flush-denormals-to-zero.  Further, check that we reflect the presence
// or absence of -fcuda-flush-denormals-to-zero in a module flag.

// CHECK-LABEL: define void @foo() #0
extern "C" __device__ void foo() {}

// FTZ: attributes #0 = {{.*}} "nvptx-f32ftz"="true"
// NOFTZ-NOT: attributes #0 = {{.*}} "nvptx-f32ftz"

// FTZ:!llvm.module.flags = !{{{.*}}[[MODFLAG:![0-9]+]]}
// FTZ:[[MODFLAG]] = !{i32 4, !"nvvm-reflect-ftz", i32 1}

// NOFTZ:!llvm.module.flags = !{{{.*}}[[MODFLAG:![0-9]+]]}
// NOFTZ:[[MODFLAG]] = !{i32 4, !"nvvm-reflect-ftz", i32 0}