diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2008-04-05 00:30:36 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2008-04-05 00:30:36 +0000 |
| commit | 0c0f83ff5d214a7f42e86ae62814526ba40a28cf (patch) | |
| tree | ffb95313874d040042271f886170c24b80ebd693 /test/CodeGen/X86/vec_shuffle-16.ll | |
| parent | a4091d34f3c05717eb5bb66a1257a0c1005e6dfa (diff) | |
| download | external_llvm-0c0f83ff5d214a7f42e86ae62814526ba40a28cf.zip external_llvm-0c0f83ff5d214a7f42e86ae62814526ba40a28cf.tar.gz external_llvm-0c0f83ff5d214a7f42e86ae62814526ba40a28cf.tar.bz2 | |
Favors pshufd over shufps when shuffling elements from one vector. pshufd is faster than shufps.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49244 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/vec_shuffle-16.ll')
| -rw-r--r-- | test/CodeGen/X86/vec_shuffle-16.ll | 25 |
1 files changed, 25 insertions, 0 deletions
diff --git a/test/CodeGen/X86/vec_shuffle-16.ll b/test/CodeGen/X86/vec_shuffle-16.ll new file mode 100644 index 0000000..2e12f6e --- /dev/null +++ b/test/CodeGen/X86/vec_shuffle-16.ll @@ -0,0 +1,25 @@ +; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse | grep shufps | count 4 +; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse | grep mov | count 2 +; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | grep pshufd | count 4 +; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | not grep shufps +; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | not grep mov + +define <4 x float> @t1(<4 x float> %a, <4 x float> %b) nounwind { + %tmp1 = shufflevector <4 x float> %b, <4 x float> undef, <4 x i32> zeroinitializer + ret <4 x float> %tmp1 +} + +define <4 x float> @t2(<4 x float> %A, <4 x float> %B) nounwind { + %tmp = shufflevector <4 x float> %A, <4 x float> %B, <4 x i32> < i32 3, i32 3, i32 3, i32 3 > + ret <4 x float> %tmp +} + +define <4 x float> @t3(<4 x float> %A, <4 x float> %B) nounwind { + %tmp = shufflevector <4 x float> %A, <4 x float> %B, <4 x i32> < i32 4, i32 4, i32 4, i32 4 > + ret <4 x float> %tmp +} + +define <4 x float> @t4(<4 x float> %A, <4 x float> %B) nounwind { + %tmp = shufflevector <4 x float> %A, <4 x float> %B, <4 x i32> < i32 1, i32 3, i32 2, i32 0 > + ret <4 x float> %tmp +} |