[GHC] #5208: Unroll array copy/clone primops in the native and LLVM code generators

[GHC]

#5208: Unroll array copy/clone primops in the native and LLVM code generators
---------------------------------+------------------------------------------
    Reporter:  tibbe             |       Owner:              
        Type:  feature request   |      Status:  new         
    Priority:  normal            |   Component:  Compiler    
     Version:  7.0.3             |    Keywords:              
    Testcase:                    |   Blockedby:              
          Os:  Unknown/Multiple  |    Blocking:              
Architecture:  Unknown/Multiple  |     Failure:  None/Unknown
---------------------------------+------------------------------------------
 #4928 added inline primops for copying and cloning arrays. It would be
 beneficial to create an unrolled loop, instead of a call to `memcpy`, when
 the copy is of a statically know small size. I suggest we use an unrolling
 threshold of 128 bytes, as that's what both GCC and LLVM use for the same
 optimization.

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5208>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler