diff options
| author | Andrea Di Biagio <Andrea_DiBiagio@sn.scee.net> | 2014-11-24 12:23:15 +0000 |
|---|---|---|
| committer | Andrea Di Biagio <Andrea_DiBiagio@sn.scee.net> | 2014-11-24 12:23:15 +0000 |
| commit | 23e2cfa834adc8a22e5d3aa3772888596d632896 (patch) | |
| tree | 846163738cdd8762f0abc9b8b30c70f85014c579 /llvm/lib/CodeGen/MachinePassRegistry.cpp | |
| parent | da50ff8e4a63a1bb0f80b1c1f3d71db84485fd77 (diff) | |
| download | bcm5719-llvm-23e2cfa834adc8a22e5d3aa3772888596d632896.tar.gz bcm5719-llvm-23e2cfa834adc8a22e5d3aa3772888596d632896.zip | |
[X86] Improved target specific combine on VSELECT dag nodes.
This patch teaches function 'transformVSELECTtoBlendVECTOR_SHUFFLE' how to
convert VSELECT dag nodes to shuffles on targets that do not have SSE4.1.
On pre-SSE4.1 targets, we can still perform blend operations using movss/movsd.
Also, removed a target specific combine that performed a premature lowering of
VSELECT nodes to target specific MOVSS/MOVSD nodes.
llvm-svn: 222647
Diffstat (limited to 'llvm/lib/CodeGen/MachinePassRegistry.cpp')
0 files changed, 0 insertions, 0 deletions
