wangtao
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Reapply last harmless part of r112934, the pattern fragment to match X86Unpcklpd 

llvm-svn: 113009
This commit is contained in:
Bruno Cardoso Lopes 2010-09-03 20:44:26 +00:00
parent fe8717c573
commit a750d994fe
1 changed files with 7 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
llvm/lib/Target/X86/X86InstrSSE.td
View File

@ -5819,6 +5819,13 @@ def : Pat<(v2i64 (X86Movlhps VR128:$src1, VR128:$src2)),
def : Pat<(v2f64 (X86Movlhpd VR128:$src1,
(scalar_to_vector (loadf64 addr:$src2)))),
(MOVHPDrm VR128:$src1, addr:$src2)>;
// FIXME: Instead of X86Unpcklpd, there should be a X86Movlhpd here, the problem
// is during lowering, where it's not possible to recognize the load fold cause
// it has two uses through a bitcast. One use disappears at isel time and the
// fold opportunity reappears.
def : Pat<(v2f64 (X86Unpcklpd VR128:$src1,
(scalar_to_vector (loadf64 addr:$src2)))),
(MOVHPDrm VR128:$src1, addr:$src2)>;
// Shuffle with MOVSS
def : Pat<(v4f32 (X86Movss VR128:$src1, (scalar_to_vector FR32:$src2))),