[FastISel][X86] Optimize predicates and fold CMP instructions.
authorJuergen Ributzka <juergen@apple.com>
Tue, 17 Jun 2014 21:55:43 +0000 (21:55 +0000)
committerJuergen Ributzka <juergen@apple.com>
Tue, 17 Jun 2014 21:55:43 +0000 (21:55 +0000)
commite8cb2ee1cd8ad7568a81f294ba09e0af2aff9e77
tree8e56670afbde026bf8a02dec1892b4cda64db5ed
parent97a66e604970685e7c7c954d3fb9ab6dab2c6f8e
[FastISel][X86] Optimize predicates and fold CMP instructions.

This optimizes predicates for certain compares, such as fcmp oeq %x, %x to
fcmp ord %x, %x. The latter one is more efficient to generate.

The same optimization is applied to conditional branches.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@211126 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86FastISel.cpp
test/CodeGen/X86/fast-isel-cmp-branch3.ll [new file with mode: 0644]
test/CodeGen/X86/fast-isel-cmp.ll