GHC 7.8.3 release

Manuel M T Chakravarty chak at cse.unsw.edu.au
Wed May 28 02:17:01 UTC 2014


Mark, 

Did you see this thread?

Manuel

Yitzchak Gale <gale at sefer.org>:

> OK then. It's just a question of updating some scripts,
> rebuilding and retesting. If these are serious generable
> usability issues, you are right that it sounds compelling.
> Still, it will add time, so Mark will have to decide.
> 
> Thanks,
> Yitz
> 
> On Tue, May 27, 2014 at 2:29 PM, Herbert Valerio Riedel
> <hvriedel at gmail.com> wrote:
>> On 2014-05-27 at 12:47:53 +0200, Yitzchak Gale wrote:
>>> Herbert Valerio Riedel wrote:
>>>> What kind of version tweaks are you thinking about? (Afaics, the GHC
>>>> bundled libraries don't have any version bumps.)
>>> 
>>> Well for one thing, wouldn't the version of base be bumped?
>> 
>> There's not been any change yet to base in the ghc-7.8 branch since the
>> 7.8.2 release:
>> 
>>  http://git.haskell.org/packages/base.git/shortlog/refs/heads/ghc-7.8
>> 
>> as you can see from
>> 
>>  https://ghc.haskell.org/trac/ghc/wiki/Commentary/Libraries/VersionHistory
>> 
>> it's happened in the past already, that 'base' was not bumped for every
>> (minor) GHC release. It's rather the exception when `base` version
>> bumps are performed for minor GHC releases.
>> 
>>> If the bugs fixed in 7.8.3 affect any packages in the platform,
>>> we would want to make whatever changes are needed to make sure
>>> that the fixes work for them - at least bumping base.
>> 
>> While I appreciate there may be packages affected by GHC 7.8.2/7.8.3
>> changes, I still fail to see how artificially bumping base would help
>> here.
>> 
>>> Then everything would need to be re-built and re-tested. And if no
>>> packages in the platform are affected, then given the ridiculous delay
>>> of the platform release already, it's harder justify waiting any
>>> longer. Like I said - it would be a difficult decision (and not mine
>>> to make).
>> 
> _______________________________________________
> ghc-devs mailing list
> ghc-devs at haskell.org
> http://www.haskell.org/mailman/listinfo/ghc-devs



More information about the ghc-devs mailing list