> Hmm I think this one belongs on the ghc trac.
> The version doesn't seem to exist in master either
> https://github.com/ghc/ghc/commits/master/libraries/ghc-prim/ghc-prim.cabal > and only exists in the 8.2 branch.
> The commit and bug report it references #14171 doesn't seem to have any
> code changes to ghc-prim (that I can see). So I'm not sure why this was
The change in that commit fixes the semantics of retry# which is
exported from GHC.Prim, so I felt that a minor bump at very least was
> So this version is a bit of a mystery..
> On Tue, Dec 19, 2017, 20:47 Michael Snoyman <[hidden email]> wrote:
>> The Stackage issue tracker just received an issue about ghc-prim 0.5.1.1.
>> https://github.com/fpco/stackage/issues/3115 >>
>> This version of the package has not been uploaded to Hackage, so:
>> * The Stackage snapshots refer to 0.5.1.1
>> * The stackage.org documentation shows version 0.5.1.1
>> * `stack unpack` and `cabal unpack` both fail on ghc-prim-0.5.1.1
>> Also, it appears that the GHC 8.2.2 release notes still refer to ghc-prim
Indeed, keeping this in sync has always been a nightmare. I've reworked
the release notes in 8.4 to make this correct by construction.
>> I'm happy to file an issue for this, but last time this popped up it was
>> unclear if the issue should be filed on GHC Trac or the Hackage Trustees
>> issue tracker.
Herbert says he will upload ghc-prim in the next day or so.