GitLab

This shouldn't be any slower; the FilePaths were already going via
String so that we could normalise them.

This means that very early on, when printing flag parsing errors, we
can do so without having a DynFlags yet.

A side-effect is that we can no longer use the LogAction in
defaultErrorHandler, as we don't have DynFlags at that point.
But all that defaultErrorHandler did is to print Strings as
SevFatal, so now it takes a 'FatalMessager' instead.

We don't use it yet

We now include the String and the SDoc in the exception, and don't
flatten them into a String until near the top-level

simplifyTop, code beautification etc. Important things:

(a) New top-level defaulting plan, gotten rid of the SimplContext field.
    See Note [Top-level Defaulting Plan]

(b) Serious bug fix in the floatEqualities mechanism
    See Note [Extra TcS Untouchables],[Float Equalities out of Implications]

The changes are mostly confined in TcSimplify but there is a
simplification wave affecting other modules as well.

with functional dependencies, and interactions with the top-level,
as well as minor refactoring.

Fixes Trac #6147

See Note [The ambiguity check for type signatures] in TcMType,
and Trac #6134, which this change fixes.

A bit of refactoring as usual.

A thunk with no free variables was not getting blackholed when
-feager-blackholing was on, but we were nevertheless pushing the
stg_bh_upd_frame version of the update frame that expects to see a
black hole.

I fixed this twice for good measure:

 - we now call blackHoleOnEntry when pushing the update frame to check
   whether the closure was actually blackholed, and so that we use the
   same predicate in both places

 - we now black hole thunks even if they have no free variables.
   These only occur when optimisation is off, but presumably if you say
   -feager-blackholing then that's what you want to happen.

If we are interrupted to do a GC, then we do not immediately do another
one.  This avoids a starvation situation where one Capability keeps
forcing a GC and the other Capabilities make no progress at all.

Previously (Trac #6148) we were only complaining for the
distfix syntax (a,b,c).

This is (I hope) the last major patch for kind polymorphism.
The big new feature is polymorphic kind recursion when you
supply a complete kind signature for a type constructor.
(I've documented it in the user manual too.)

This fixes Trac #6137, #6093, #6049.

The patch also makes type/data families less polymorphic by default.
   data family T a
now defaults to T :: * -> *
If you want T :: forall k. k -> *, use
   data family T (a :: k)

This defaulting to * is done whenever there is a
"complete, user-specified kind signature", something that is
carefully defined in the user manual.

Hurrah!

The bindist install directory was wrong.

Make InteractiveEval.setContext throw a clearer exception when it is
asked to add an IIModule which is not a home module or is not
interpreted.