diff options
| author | Nadav Rotem <nadav.rotem@intel.com> | 2011-03-19 13:09:10 +0000 |
|---|---|---|
| committer | Nadav Rotem <nadav.rotem@intel.com> | 2011-03-19 13:09:10 +0000 |
| commit | 06cc324b9da1dc8fb7360a560343c28f5e7a940a (patch) | |
| tree | f055ce6e171c709584dc15a25174178955a37bbf /test/CodeGen | |
| parent | 415247dac4cdbec704af6caa69aff2ac63f21377 (diff) | |
| download | external_llvm-06cc324b9da1dc8fb7360a560343c28f5e7a940a.zip external_llvm-06cc324b9da1dc8fb7360a560343c28f5e7a940a.tar.gz external_llvm-06cc324b9da1dc8fb7360a560343c28f5e7a940a.tar.bz2 | |
Add support for legalizing UINT_TO_FP of vectors on platforms which do
not have native support for this operation (such as X86).
The legalized code uses two vector INT_TO_FP operations and is faster
than scalarizing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127951 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen')
| -rw-r--r-- | test/CodeGen/X86/vec_uint_to_fp.ll | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/test/CodeGen/X86/vec_uint_to_fp.ll b/test/CodeGen/X86/vec_uint_to_fp.ll new file mode 100644 index 0000000..39e7d71 --- /dev/null +++ b/test/CodeGen/X86/vec_uint_to_fp.ll @@ -0,0 +1,11 @@ +; RUN: llc < %s -march=x86 -mcpu=sandybridge | FileCheck %s + +; Test that we are not lowering uinttofp to scalars +define <4 x float> @test1(<4 x i32> %A) nounwind { +; CHECK: test1: +; CHECK-NOT: cvtsd2ss +; CHECK: ret + %C = uitofp <4 x i32> %A to <4 x float> + ret <4 x float> %C +} + |