This showed up when converting ds057 to follow the new bang pattern rules,
in #2806.

For now we only get a warning, rather than an error, because the alex
and happy templates don't follow the new rules yet.

The top-level indirection-shorting, done in SimplCore.shortOutIndirections,
was accidentally nuking a NOINLINE pragma.  This patch adopts a slightly
more conservative approach to indirection-shorting, only doing so if 
we do not thereby overwrite or lose some user-supplied pragmas.

   movss instruction was being used to move the result into the 
   destination register instead of movsd

Also some tidyups and renaming

Part of the fix for #3171

Part of the fix for #3171

When making record selectors into ordinary functions (rather than
magial implicit Ids) I forgot that they could therefore show up in 
the *local* TcEnv, not the global one.  This fixes that problem,
and thereby Trac #3176.

In GHCi if you say 'import Foo' meaning to load a package module Foo,
and Foo.hs is found on the search path, then GHCi replies "module Foo
is not loaded", because it knows Foo refers to the source file rather
than the package module, and you haven't loaded that module with
:load.

This is consistent with the usual module-finding semantics.  However,
it isn't particularly useful.  And it leads to silly problems like not
being able to start GHCi when you happen to be sitting in
libraries/base, because GHCi thinks the Prelude hasn't been loaded.

So now I've made a slight change to the way that 'import M' works: if
M is loaded, then it refers to the loaded module, otherwise it looks
for a package module M.  This does what the reporter of #2682 wanted,
and since it turns an error condition into meaningful behaviour it
can't break anything.  

The only undesirable consequence is that 'import M' might refer to a
different M than ':load M'.  Hopefully that won't lead to confusion.

It turns out we can easily support breakpoints on expressions with
unlifted types, by translating 

  case tick# of _ -> e

into

  let f = \s . case tick# of _ -> e 
  in  f realWorld#

instead of just a plain let-binding.  This is the same trick that GHC
uses for abstracting join points of unlifted type.

In #2845, GHC has eta-expanded the tick expression, changing the
result type from IO a to (# State#, a #), which was the reason the
tick was suddenly being ignored.  By supporting ticks on unlifted
expressions we can make it work again, although some confusion might
arise because _result will no longer be available (it now has
unboxed-tuple type, so we can't bind it in the environment).  The
underlying problem here is that GHC does transformations like
eta-expanding the tick expressions, and there's nothing we can do to
prevent that.

Can't use windowed regs because the window moves during a function
call. Can't use the global regs because they're reserved for other purposes.

My previous patch didn't completely solve the problem.
I believe I got it right this time.

Fix a bug in :print affecting data types with unboxed components

  The x86 code generator sometimes generates nop moves like
  movl %vI_a %vI_a, but we'll just ignore them for coalescing
  purposes instead of emitting an error. It shouldn't hurt anything.

I forgot to record some additional changes.

Simon P-J suggested the following simplifications in #3097:

* Allow nested foralls in `ctype` just like in `ctypedoc`.
* Use `gentype` rather than `type` in the LHS of type declarations.
* Inline `type` in `ctype`.
* Rename `gentype` to `type`.

This patch does this. Also, the equivalent thing is done for documented types.

postEventTypeID was used to post event marker IDs, but instead we just
call postWord16 directly; thus, this function is unused.

For consistency, changed access of thread id to be through capability instead of directly from StgRegTable.

StgWord64 other better represents this argument since it can be used as
any particular data, thus "other" and not the "from" capability as it was
previously strictly used.  Also, StgWord64 is normally larger than type
nat to allow for larger data to be passed through the "other" argument.

Fixed ThreadID to be defined as StgThreadID, not StgWord64.  Changed CapabilityNum to CapNo.  Added helper functions postCapNo() and postThreadID().
ThreadID was StgWord64, but should have been StgThreadID, which is
currently StgWord32.  Changed name from CapabilityNum to CapNo to better
reflect naming in Capability struct where "no" is the capability number.
Modified EventLog.c to use the helper functions postCapNo() and
postThreadID () for CapNo and ThreadID.

EventTypeDescriptions order must be synchronized with the event type
definitions in includes/EventLogFormat.h for the definitions to correctly
index with the matching description.

This patch sligtly re-adjusts the way in which the syntax of types 
is handled:

 * In the lexer, '.' and '*' are always accepted in types
   (previously it was conditional).  This things can't mean
   anything else in H98, which is the only reason for doing things
   conditionally in the lexer.

 * As a result '.' in types is never treated as an operator.
   Instead, lacking a 'forall' keyword, it turns into a plain parse error.

 * Test for -XKindSignatures in the renamer when processing
     a) type variable bindings
     b) types with sigs (ty :: kind-sig)

 * Make -XKindSignatures be implied by -XTypeFamilies 
   Previously this was buried in the conditonal lexing of '*'