diff options
| author | Evan Cheng <evan.cheng@apple.com> | 2010-02-16 21:09:44 +0000 |
|---|---|---|
| committer | Evan Cheng <evan.cheng@apple.com> | 2010-02-16 21:09:44 +0000 |
| commit | ae3ecf96035165de3e5327fb33bd30504e21832f (patch) | |
| tree | c1bbc762157021b71c7efd2ee65d4f4140e16fe1 /test/CodeGen/X86/sink-hoist.ll | |
| parent | 6417171026447cde57330114e7df2a22bebfc135 (diff) | |
| download | external_llvm-ae3ecf96035165de3e5327fb33bd30504e21832f.zip external_llvm-ae3ecf96035165de3e5327fb33bd30504e21832f.tar.gz external_llvm-ae3ecf96035165de3e5327fb33bd30504e21832f.tar.bz2 | |
Look for SSE and instructions of this form: (and x, (build_vector c1,c2,c3,c4)).
If there exists a use of a build_vector that's the bitwise complement of the mask,
then transform the node to
(and (xor x, (build_vector -1,-1,-1,-1)), (build_vector ~c1,~c2,~c3,~c4)).
Since this transformation is only useful when 1) the given build_vector will
become a load from constpool, and 2) (and (xor x -1), y) matches to a single
instruction, I decided this is appropriate as a x86 specific transformation.
rdar://7323335
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@96389 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/sink-hoist.ll')
| -rw-r--r-- | test/CodeGen/X86/sink-hoist.ll | 1 |
1 files changed, 0 insertions, 1 deletions
diff --git a/test/CodeGen/X86/sink-hoist.ll b/test/CodeGen/X86/sink-hoist.ll index 01d7373..e1d0fe7 100644 --- a/test/CodeGen/X86/sink-hoist.ll +++ b/test/CodeGen/X86/sink-hoist.ll @@ -63,7 +63,6 @@ entry: ; CHECK: vv: ; CHECK: LCPI4_0(%rip), %xmm0 ; CHECK: LCPI4_1(%rip), %xmm1 -; CHECK: LCPI4_2(%rip), %xmm2 ; CHECK: align ; CHECK-NOT: LCPI ; CHECK: ret |