projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8ae0f8d) | patch
AArch64: Better codegen for loading from __fp16.
authorJim Grosbach <grosbach@apple.com>
Tue, 8 Jul 2014 23:28:48 +0000 (23:28 +0000)
committerJim Grosbach <grosbach@apple.com>
Tue, 8 Jul 2014 23:28:48 +0000 (23:28 +0000)
commit04691a530d9068b5fc9bfbe7d7f157e11bdef116
tree5265d454ce18bc87e734c21c7c9b24e57fc8e396tree | snapshot
parent8ae0f8d618b9996ccd650d9339255bc67d854531commit | diff
AArch64: Better codegen for loading from __fp16.

Loading will generally extend to an f32 or an 64, so make sure
to match those patterns directly to load into the FPR16 register
class directly rather than going through the integer GPRs.

This also eliminates an extra step in the convert-to-f64 path
which was first converting to f32 and then to f64 from there.

rdar://17594379

llvm-svn: 212573
llvm/lib/Target/AArch64/AArch64InstrInfo.td diff | blob | history
llvm/test/CodeGen/AArch64/f16-convert.ll [new file with mode: 0644] blob
Domain: System / Toolchain;