projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 32d0d38) | patch
[InstCombine] make icmp-mul fold more efficient
authorSanjay Patel <spatel@rotateright.com>
Thu, 25 May 2017 14:13:57 +0000 (14:13 +0000)
committerSanjay Patel <spatel@rotateright.com>
Thu, 25 May 2017 14:13:57 +0000 (14:13 +0000)
commit5150612012beab0065080fce0507755a7e325099
tree5d003c75bdb9bf4e08e8304ffddedf07797f3260tree | snapshot
parent32d0d3867949e1c0c00e76f440a2b67f8b09c7b0commit | diff
[InstCombine] make icmp-mul fold more efficient

There's probably a lot more like this (see also comments in D33338 about responsibility),
but I suspect we don't usually get a visible manifestation.

Given the recent interest in improving InstCombine efficiency, another potential micro-opt
that could be repeated several times in this function: morph the existing icmp pred/operands
instead of creating a new instruction.

llvm-svn: 303860
llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp diff | blob | history
llvm/test/Transforms/InstCombine/icmp.ll diff | blob | history
Domain: System / Toolchain;