diff options
| author | Chris Lattner <sabre@nondot.org> | 2008-03-08 23:43:36 +0000 |
|---|---|---|
| committer | Chris Lattner <sabre@nondot.org> | 2008-03-08 23:43:36 +0000 |
| commit | f3ba434781daa1c2b16f77a91e5209c21ce12428 (patch) | |
| tree | 798ccad01fcf3bf6f276755c00f7d53ce4cf751f /test/CodeGen/X86/vec_set-9.ll | |
| parent | 19f7969f816000cf5ddf8a19163193b5feb9bc0a (diff) | |
| download | llvm-f3ba434781daa1c2b16f77a91e5209c21ce12428.tar.gz llvm-f3ba434781daa1c2b16f77a91e5209c21ce12428.tar.bz2 llvm-f3ba434781daa1c2b16f77a91e5209c21ce12428.tar.xz | |
Teach SD some vector identities, allowing us to compile vec_set-9 into:
_test3:
movd %rdi, %xmm1
#IMPLICIT_DEF %xmm0
punpcklqdq %xmm1, %xmm0
ret
instead of:
_test3:
#IMPLICIT_DEF %rax
movd %rax, %xmm0
movd %rdi, %xmm1
punpcklqdq %xmm1, %xmm0
ret
This is still not ideal. There is no reason to two xmm regs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48058 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/vec_set-9.ll')
| -rw-r--r-- | test/CodeGen/X86/vec_set-9.ll | 8 |
1 files changed, 8 insertions, 0 deletions
diff --git a/test/CodeGen/X86/vec_set-9.ll b/test/CodeGen/X86/vec_set-9.ll new file mode 100644 index 0000000000..d00ceff540 --- /dev/null +++ b/test/CodeGen/X86/vec_set-9.ll @@ -0,0 +1,8 @@ +; RUN: llvm-as < %s | llc -march=x86-64 | grep movd | count 1 + +define <2 x i64> @test3(i64 %A) { +entry: + %B = insertelement <2 x i64> undef, i64 %A, i32 1 + ret <2 x i64> %B +} + |