[llvm-commits] [llvm] r139009 - /llvm/trunk/lib/Transforms/Scalar/LoopUnrollPass.cpp

[Andrew Trick]

On Sep 5, 2011, at 11:14 AM, Nick Lewycky wrote:
>> x86 is break even.
>> ARM performance changes to expect (+ is good):
>> External/SPEC/CFP2000/183.equake/183.equake +13%
>> SingleSource/Benchmarks/Dhrystone/fldry     +21%
>> MultiSource/Applications/spiff/spiff         +3%
>> SingleSource/Benchmarks/Stanford/Puzzle     -14%
>> 
>> The Puzzle regression is actually an improvement in loop optimization
>> that defeats GVN: rdar://problem/10065079.
> 
> PR, please?

See PR10872.

-Andy