File: cuda-march.cu

package info (click to toggle)
llvm-toolchain-11 1%3A11.0.1-2
  • links: PTS, VCS
  • area: main
  • in suites: bullseye
  • size: 995,808 kB
  • sloc: cpp: 4,767,656; ansic: 760,916; asm: 477,436; python: 170,940; objc: 69,804; lisp: 29,914; sh: 23,855; f90: 18,173; pascal: 7,551; perl: 7,471; ml: 5,603; awk: 3,489; makefile: 2,573; xml: 915; cs: 573; fortran: 503; javascript: 452
file content (22 lines) | stat: -rw-r--r-- 772 bytes parent folder | download | duplicates (15) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
 
// Checks that cuda compilation does the right thing when passed -march.
// (Specifically, we want to pass it to host compilation, but not to device
// compilation or ptxas!)
//
// REQUIRES: clang-driver
// REQUIRES: x86-registered-target
// REQUIRES: nvptx-registered-target

// RUN: %clang -no-canonical-prefixes -### -target x86_64-linux-gnu -c \
// RUN: -march=haswell %s 2>&1 | FileCheck %s
// RUN: %clang -no-canonical-prefixes -### -target x86_64-linux-gnu -c \
// RUN: -march=haswell --cuda-gpu-arch=sm_20 %s 2>&1 | FileCheck %s

// CHECK: {{.*}}clang{{.*}}" "-cc1"
// CHECK-SAME: "-triple" "nvptx
// CHECK-SAME: "-target-cpu" "sm_20"

// CHECK: ptxas
// CHECK-SAME: "--gpu-name" "sm_20"

// CHECK: {{.*}}clang{{.*}}" "-cc1"
// CHECK-SAME: "-target-cpu" "haswell"