diff options
| author | Nadav Rotem <nadav.rotem@intel.com> | 2012-07-10 13:25:08 +0000 |
|---|---|---|
| committer | Nadav Rotem <nadav.rotem@intel.com> | 2012-07-10 13:25:08 +0000 |
| commit | d908ddc186aa026164104adcd4207436fe54e279 (patch) | |
| tree | d31b189018bcb84f14f4fd16a400d5682b266304 /llvm/test/CodeGen/X86/vec_compare-2.ll | |
| parent | 65fd0e1fd3c7fe24bb09e654d228c9584c2fa6ff (diff) | |
| download | bcm5719-llvm-d908ddc186aa026164104adcd4207436fe54e279.tar.gz bcm5719-llvm-d908ddc186aa026164104adcd4207436fe54e279.zip | |
Improve the loading of load-anyext vectors by allowing the codegen to load
multiple scalars and insert them into a vector. Next, we shuffle the elements
into the correct places, as before.
Also fix a small dagcombine bug in SimplifyBinOpWithSameOpcodeHands, when the
migration of bitcasts happened too late in the SelectionDAG process.
llvm-svn: 159991
Diffstat (limited to 'llvm/test/CodeGen/X86/vec_compare-2.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/vec_compare-2.ll | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/llvm/test/CodeGen/X86/vec_compare-2.ll b/llvm/test/CodeGen/X86/vec_compare-2.ll index 91777f7aa6b..46d6a23554f 100644 --- a/llvm/test/CodeGen/X86/vec_compare-2.ll +++ b/llvm/test/CodeGen/X86/vec_compare-2.ll @@ -10,8 +10,7 @@ define void @blackDespeckle_wrapper(i8** %args_list, i64* %gtid, i64 %xend) { entry: ; CHECK: cfi_def_cfa_offset ; CHECK-NOT: set -; CHECK: movzwl -; CHECK: movzwl +; CHECK: punpcklwd ; CHECK: pshufd ; CHECK: pshufb %shr.i = ashr <4 x i32> zeroinitializer, <i32 3, i32 3, i32 3, i32 3> ; <<4 x i32>> [#uses=1] |
