[GHC] #8704: Use GHC.Exts.build in randoms, randomRs to achieve fusion

[Ryan Newton]

Thanks, no I was not CC'd.  Now I've added myself  to ghc-tickets
(alongside libraries, ghc-devs, haskell-cafe ... whew ;-) .)

An initial inspection of the patch looks fine.  I'll go ahead and merge it.

However, I remind everyone that the priority should be replacing random
entirely.  It's slow and known to be a bad algorithm (for the splitting
bit).  Koen Classen & co had an implementation in the Haskell symposium
paper that may be a candidate.




On Tue, Feb 4, 2014 at 3:04 AM, Simon Peyton Jones <simonpj at microsoft.com>wrote:

> Hi Ryan
>
> Are you on this thread?   You probably should be!
>
> Simon
>
> | -----Original Message-----
> | From: ghc-tickets [mailto:ghc-tickets-bounces at haskell.org] On Behalf Of
> | GHC
> | Sent: 27 January 2014 18:49
> | Cc: ghc-tickets at haskell.org
> | Subject: Re: [GHC] #8704: Use GHC.Exts.build in randoms, randomRs to
> | achieve fusion
> |
> | #8704: Use GHC.Exts.build in randoms, randomRs to achieve fusion
> | -------------------------------------+---------------------------------
> | ---
> |         Reporter:  ion1              |            Owner:
> |             Type:  feature request   |           Status:  patch
> |         Priority:  normal            |        Milestone:
> |        Component:  libraries/random  |          Version:  7.6.3
> |       Resolution:                    |         Keywords:  fusion
> | Operating System:  Unknown/Multiple  |     Architecture:
> | Unknown/Multiple
> |  Type of failure:  None/Unknown      |       Difficulty:  Unknown
> |        Test Case:                    |       Blocked By:
> |         Blocking:                    |  Related Tickets:  #4218
> | -------------------------------------+---------------------------------
> | ---
> |
> | Comment (by nomeata):
> |
> |  Thanks. From my POV it is worth adding even if you can't measure a
> |  performance gain; it is still good to know that nice code is being
> |  generated. But of course it is up to Ryan Newton (random maintainer)
> | to
> |  decide this.
> |
> | --
> | Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8704#comment:4>
> | GHC <http://www.haskell.org/ghc/>
> | The Glasgow Haskell Compiler
> | _______________________________________________
> | ghc-tickets mailing list
> | ghc-tickets at haskell.org
> | http://www.haskell.org/mailman/listinfo/ghc-tickets
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.haskell.org/pipermail/ghc-devs/attachments/20140204/6d2a8283/attachment.html>