[GHC] #3474: Add a strict variant of iterate to Data.List

GHC ghc-devs at haskell.org
Wed Aug 16 11:54:13 UTC 2017


#3474: Add a strict variant of iterate to Data.List
-------------------------------------+-------------------------------------
        Reporter:  mux               |                Owner:  (none)
            Type:  proposal          |               Status:  new
        Priority:  normal            |            Milestone:  8.2.2
       Component:  libraries/base    |              Version:  6.10.4
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Changes (by winter):

 * status:  closed => new
 * resolution:  wontfix =>
 * milestone:  Not GHC => 8.2.2


Comment:

 Today a friend asked me the exactly problem, i.e. why `iterate (+1) 1 !!
 100000000` cost 8G ram. This is surprising because most people are
 expecting GC should chase the list and the whole computation should run in
 O(1) space.

 But as soon as i checked the definition of `iterate`, i realize it's no
 possible to run in O(1) space, we're simply not forcing the `f (f ... (f
 x))` chain at all.

 So i came to this ticket. Adding `iterate'` may not be enough: we have to
 add document to notify people to watch out!

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/3474#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list