projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8d1b7e5) | patch
Add Neon VCVT instructions for f32 <-> f16 conversions.
authorBob Wilson <bob.wilson@apple.com>
Wed, 15 Dec 2010 22:14:12 +0000 (22:14 +0000)
committerBob Wilson <bob.wilson@apple.com>
Wed, 15 Dec 2010 22:14:12 +0000 (22:14 +0000)
commit0406356cd4cb7b689e2472faa8dfb7d721f9d274
tree0c1e2f26438acad7f4053004cd38fe2d9282c52dtree | snapshot
parent8d1b7e57e56015576fd489a57d74c53b98c5a56fcommit | diff
Add Neon VCVT instructions for f32 <-> f16 conversions.
Clang is now providing intrinsics for these and so we need to support them
in the backend.  Radar 8068427.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121902 91177308-0d34-0410-b5e6-96231b3b80d8
include/llvm/IntrinsicsARM.td diff | blob | history
lib/Target/ARM/ARM.td diff | blob | history
lib/Target/ARM/ARMInstrInfo.td diff | blob | history
lib/Target/ARM/ARMInstrNEON.td diff | blob | history
test/CodeGen/ARM/vcvt.ll diff | blob | history
test/MC/ARM/neon-convert-encoding.s diff | blob | history
test/MC/ARM/neont2-convert-encoding.s diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors