Contradiction regarding Foreign.C.Types and marshable foreign types

SPJ writes in #GHC9079:

Well, this looks to me like a bug in the Haskell 2010 report

* Section 8.4.2 says that newtypes must have their constructors visible, and claims that Foreign.C.Types follows this convention.

* But Chapter 28 which shows Foreign.C.Types manifestly does not follow this convention.

That looks like an outright contradiction.

I think we should export all the data constructors in Foreign.C.Types in the haskell2010 package. Does anyone disagree?

Trac metadata

Trac field	Value
Version	2010
Type	Bug
TypeOfFailure	OtherFailure
Priority	high
Resolution	Unresolved
Component
Test case
Differential revisions
BlockedBy
Related
Blocking
CC
Operating system
Architecture