vector for DPH

Manuel M T Chakravarty chak at cse.unsw.edu.au
Tue Sep 21 22:02:38 EDT 2010


Duncan Coutts:
> On 22 September 2010 02:05Duncan Coutts <duncan at well-typed.com> wrote:
>> On 22 September 2010 01:32, Manuel M T Chakravarty <chak at cse.unsw.edu.au> wrote:
>> 
>>>> Manuel: would you like to talk to Ben, Gabi, Roman, and decide what you'd like us to do for the release.  Specifically:
>>>> 
>>>> - Release with DPH, or rely on 'cabal install dph'?
>>>> - Push the new refactoring into the branch; or not?
>>> 
>>> Below I attach a summary of our Tuesday Skype call.  I would like to stick to the plan that we formulated during that call — that is:
>>> 
>>> * Release 7.0 including DPH using the new library that is based on vector.
>> 
>> Can I make another suggestion for a way to let ghc ship vector etc
>> without needing to put vector through the HP package proposal process:
>> put all the dph "backend" libs into a separate package database and
>> make -fvectorise imply not just -package blah but also the appropriate
>> -package-conf flag.
> 
> Just to note for everyone else: my suggestion will not work because
> there is no single flag that distinguishes DPH programs. DPH programs
> can import dph modules but otherwise not use any special language
> features or compilation options.
> 
> As a simple alternative we discussed using a package name like
> "dph-vector", though I would prefer something more obscure still so
> that users who read Don's blog posts about how cool the vector package
> is will not end up depending on this package. We don't want to reuse
> the name "dph-prim-seq" because that'd confuse everyone. How about
> "dph-prim-vec"?

I think using a different package name is a reasonable compromise.  Personally, I think "dph-vector" or "ghc-vector" would be sufficiently explicit to show that it is not a HP sanctioned package, but I could live with something slightly more obscure, too.

Manuel



More information about the Cvs-ghc mailing list