maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/CodeGen/convergent-functions.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

9 lines
375 B
C++
Raw Normal View History

clang: Add -fconvergent-functions flag The CUDA builtin library is apparently compiled in C++ mode, so the assumption of convergent needs to be made in a typically non-SPMD language. The functions in the library should still be assumed convergent. Currently they are not, which is potentially incorrect and this happens to work after the library is linked.
2019-10-27 08:15:10 +08:00
// RUN: %clang_cc1 -triple i386-pc-win32 -emit-llvm -fconvergent-functions -o - < %s | FileCheck -check-prefix=CONVFUNC %s
// RUN: %clang_cc1 -triple i386-pc-win32 -emit-llvm -o - < %s | FileCheck -check-prefix=NOCONVFUNC %s
// Test that the -fconvergent-functions flag works
// CONVFUNC: attributes #0 = { convergent {{.*}} }
// NOCONVFUNC-NOT: convergent
void func() { }