projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 94fe5c1) | patch
AArch64: improve handling & modelling of FP_TO_XINT nodes.
authorTim Northover <tnorthover@apple.com>
Sun, 15 Jun 2014 09:27:15 +0000 (09:27 +0000)
committerTim Northover <tnorthover@apple.com>
Sun, 15 Jun 2014 09:27:15 +0000 (09:27 +0000)
commit8bfc50e4a976476425e8f2ba7e71d7d14b48b474
tree9ccc4b1232fe8d9bb994b2eeab74a8a3a9a3b7aatree | snapshot
parent94fe5c1fe2a3419d90e612735200dbc2c36f29efcommit | diff
AArch64: improve handling & modelling of FP_TO_XINT nodes.

There's probably no acatual change in behaviour here, just updating
the LowerFP_TO_INT function to be more similar to the reverse
implementation and updating costs to current CodeGen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210985 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AArch64/AArch64ISelLowering.cpp diff | blob | history
lib/Target/AArch64/AArch64TargetTransformInfo.cpp diff | blob | history
test/CodeGen/AArch64/complex-fp-to-int.ll diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors