[GHC] #16052: Core optimizations for memset on a small range

[GHC]

#16052: Core optimizations for memset on a small range
-------------------------------------+-------------------------------------
        Reporter:  andrewthad        |                Owner:  (none)
            Type:  task              |               Status:  new
        Priority:  normal            |            Milestone:  8.9
       Component:  Compiler          |              Version:  8.6.3
      Resolution:                    |             Keywords:  newcomer
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by AndreasK):

 I'm happy to take a look once there is something to look at.

 > I don’t know how the NCG normally deals with the 32 bit vs 64 bit
 distinction in other places.

 A lot of code is in the spirit of `is32Bit then foo32 else foo64`.

 > It’s probably difficult to add a better test for this once it’s fixed.
 It would need to compile starting with STG, not Cmm (which is what the
 current test does), and go all the way down to asm. But it might be
 brittle.

 Brittle in what sense? Just grepping for calls to memset should be quite
 stable as far as making sure we don't emit calls like this.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/16052#comment:7>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler