diff options
| author | Robert Lougher <rob.lougher@gmail.com> | 2016-01-12 11:48:25 +0000 | 
|---|---|---|
| committer | Robert Lougher <rob.lougher@gmail.com> | 2016-01-12 11:48:25 +0000 | 
| commit | 6abd69a60b74584e86102d400880d2ae00932f84 (patch) | |
| tree | 36936a765869d4ba9c9523a593be52a090d48cc6 /clang/lib/CodeGen/CodeGenFunction.cpp | |
| parent | 239132f452ee096d2701d7039fd148600aa37478 (diff) | |
| download | llvm-6abd69a60b74584e86102d400880d2ae00932f84.zip llvm-6abd69a60b74584e86102d400880d2ae00932f84.tar.gz llvm-6abd69a60b74584e86102d400880d2ae00932f84.tar.bz2 | |
The isel pattern that selects the memory-register form of VCVTPH2PS
(64 to 128-bit) matches against the pattern fragment 'vzmovl_v2i64'
(a zero-extended 64-bit load).
However, a change in r248784 teaches the instruction combiner that only
the lower 64 bits of the input to a 128-bit vcvtph2ps are used.  This means
the instruction combiner will ordinarily optimize away the upper 64-bit
insertelement instruction in the zero-extension and so we no longer select
the memory-register form.  To fix this a new pattern has been added.
Differential Revision: http://reviews.llvm.org/D16067
llvm-svn: 257470
Diffstat (limited to 'clang/lib/CodeGen/CodeGenFunction.cpp')
0 files changed, 0 insertions, 0 deletions
