[GHC] #13654: Optimize casMutVar# for single-threaded runtime

[GHC]

#13654: Optimize casMutVar# for single-threaded runtime
-------------------------------------+-------------------------------------
        Reporter:  dfeuer            |                Owner:  (none)
            Type:  feature request   |               Status:  new
        Priority:  normal            |            Milestone:  8.4.1
       Component:  Runtime System    |              Version:  8.2.1-rc1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Description changed by dfeuer:

@@ -1,4 +1,4 @@
- When using the non-threaded RTS, `stg_casMutVarzh`,
- `stg_atomicModifyMutVarzh`, etc., shouldn't need to actually use atomic
- instructions, but they seem to do so. I believe this makes them
- substantially slower than necessary in that context.
+ In the non-threaded RTS, `stg_casMutVarzh`, `stg_atomicModifyMutVarzh`,
+ etc., shouldn't need to actually use atomic instructions, but they seem to
+ do so. I believe this makes them  substantially slower than necessary in
+ that context.

New description:

 In the non-threaded RTS, `stg_casMutVarzh`, `stg_atomicModifyMutVarzh`,
 etc., shouldn't need to actually use atomic instructions, but they seem to
 do so. I believe this makes them  substantially slower than necessary in
 that context.

--

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13654#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler