diff options
author | Nicolai Hahnle <nhaehnle@gmail.com> | 2015-12-15 17:24:15 +0000 |
---|---|---|
committer | Nicolai Hahnle <nhaehnle@gmail.com> | 2015-12-15 17:24:15 +0000 |
commit | 78fd4f087b0832bdfef1c7db54bc234002e2ff30 (patch) | |
tree | ee4aa587893ee78d39d2ff516ba4bf67dd358d0c /llvm/lib/Analysis/TargetLibraryInfo.cpp | |
parent | 8f307217c3887087f22f3bef30bbdeeb00ee5299 (diff) | |
download | llvm-78fd4f087b0832bdfef1c7db54bc234002e2ff30.zip llvm-78fd4f087b0832bdfef1c7db54bc234002e2ff30.tar.gz llvm-78fd4f087b0832bdfef1c7db54bc234002e2ff30.tar.bz2 |
AMDGPU: mark ldexp LibCalls as unavailable
Summary:
The LibCallSimplifier will turn llvm.exp2.* intrinsics into ldexp* libcalls
which do not make sense with the AMDGPU backend.
In the long run, we'll want an llvm.ldexp.* intrinsic to properly make use of
this optimization, but this works around the problem for now.
See also: http://reviews.llvm.org/D14327 (suggested llvm.ldexp.* implementation)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92709
Reviewers: arsenm, tstellarAMD
Differential Revision: http://reviews.llvm.org/D14990
llvm-svn: 255658
Diffstat (limited to 'llvm/lib/Analysis/TargetLibraryInfo.cpp')
-rw-r--r-- | llvm/lib/Analysis/TargetLibraryInfo.cpp | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/llvm/lib/Analysis/TargetLibraryInfo.cpp b/llvm/lib/Analysis/TargetLibraryInfo.cpp index 845ca05..e00f4ae 100644 --- a/llvm/lib/Analysis/TargetLibraryInfo.cpp +++ b/llvm/lib/Analysis/TargetLibraryInfo.cpp @@ -61,6 +61,13 @@ static void initialize(TargetLibraryInfoImpl &TLI, const Triple &T, } #endif // !NDEBUG + if (T.getArch() == Triple::r600 || + T.getArch() == Triple::amdgcn) { + TLI.setUnavailable(LibFunc::ldexp); + TLI.setUnavailable(LibFunc::ldexpf); + TLI.setUnavailable(LibFunc::ldexpl); + } + // There are no library implementations of mempcy and memset for AMD gpus and // these can be difficult to lower in the backend. if (T.getArch() == Triple::r600 || |