GitLab

---------------------
	Dealing with deriving
	---------------------

I spent a ridiculously long time peering at a bug report whereby
a 'deriving' clause sent GHC 5.02.1 into a loop.  It was all to
do with allowing instances like

	instance Foo a b => Baz (T a)

(Notice the 'b' on the left which does not appear on the right.)

I realised that it's hard for the deriving machinery to find a
fixpoint when these sort of instance decls are around.  So I
now constrain *derived* instance decls not to have this form;
all the tyvars on the left must appear on the right.

On the way I commoned up the previously-separate tcSimplify
machinery for 'deriving' and 'default' decls with that for
everything else.   As a result, quite a few files are touched.

I hope I havn't broken anything.

-----------------------------
	Buglet in rank-N polymorphism
	-----------------------------

Fix a bug in the way result type signatures are handled; they
hadn't been brought into the rank-N polymorphism world.

heal

debugging-by-nightly-build-reports; give poke_c_flag a correct FI decl

Get rid of uses of #const, #peek, #poke and #ptr from PrelPosix.hsc
(this leaves just uses of #type in PrelPosix) - provide constant and
accessor wrappers via PrelIOUtils.c instead.

Who knows, we might just be able to bootstrap via .hc files again..?

Only compiled & tested under Win32.

Added & exported low-level memcpy() wrappers - possible to offset both
src and dest.

- removed PrelPosix.c_read, PrelPosix.c_write
- added Ptr and RawBuffer versions of PrelHandle.write_off,
  and PrelHandle.read_off and exported them, i.e., PrelHandle
  now exports:

    read_off_ba :: FD -> Bool -> RawBuffer -> Int -> CInt -> IO CInt
    read_off :: FD -> Bool -> Ptr CChar -> Int -> CInt -> IO CInt
    write_off_ba :: CInt -> Bool -> RawBuffer -> Int -> CInt -> IO CInt
    write_off :: CInt -> Bool -> Ptr CChar -> Int -> CInt -> IO CInt

- make hPutChar (win)socket friendly.

add exeext to HS_PROG to avoid redundant re-linking

derive Eq Cond so that an assertion in MachCode works.

wibble

Wibble

Allow record declaratoins with zero fields.

Docment newtype deriving stuff

More wibbles to newtype deriving

Move some small inline-in-C functions out of PrelPosix.hsc and into
the supporting cbits sources, where they belong.

Following this change, can you now compile the entire Prelude on
x86-linux with the native code generator, and get something which
actually works properly.

Wibbles

Comments

Now that we can derive arbitrary classes for newtypes, use this to
derive all the classes we need for the types in PrelCTypes and
PrelCTypesISO, instead of the previous hack of explicitly coercing the
methods.

This should improve the quality of code using these types too; the
coercions were surely getting in the way of optimisations before.

Outline freeStablePtr, so NCG-generated code actually has something to call :)

Also remove the presumably-defunct splitStablePtr.

Generate floating-point comparisons on x86 which deal with NaNs in what
I assume is an IEEE854 compliant fashion.  For
   == >= > <= <
if either arg is a NaN, produce False, and for
   /=
if either arg is a NaN, produce True.

This is the behaviour that gcc has, by default.

Requires some ultramagical x86 code frags to be emitted.  A big comment
in PprMach explains how it works.

nano-comment-wibble

---------------------------------------------
	More type system extensions (for John Hughes)
	---------------------------------------------

1.  Added a brand-new extension that lets you derive ARBITRARY CLASSES
for newtypes.  Thus

	newtype Age = Age Int deriving( Eq, Ord, Shape, Ix )

The idea is that the dictionary for the user-defined class Shape Age
is *identical* to that for Shape Int, so there is really no deriving
work to do.   This saves you writing the very tiresome instance decl:

	instance Shape Age where
	   shape_op1 (Age x) = shape_op1 x
	   shape_op2 (Age x1) (Age x2) = shape_op2 x1 x2
	   ...etc...

It's more efficient, too, becuase the Shape Age dictionary really
will be identical to the Shape Int dictionary.

There's an exception for Read and Show, because the derived instance
*isn't* the same.

There is a complication where higher order stuff is involved.  Here is
the example John gave:

   class StateMonad s m | m -> s where ...

   newtype Parser tok m a = Parser (State [tok] (Failure m) a)
			  deriving( Monad, StateMonad )

Then we want the derived instance decls to be

   instance Monad (State [tok] (Failure m)) => Monad (Parser tok m)
   instance StateMonad [tok] (State [tok] (Failure m))
	 => StateMonad [tok] (Parser tok m)

John is writing up manual entry for all of this, but this commit
implements it.   I think.


2.  Added -fallow-incoherent-instances, and documented it.  The idea
is that sometimes GHC is over-protective about not committing to a
particular instance, and the programmer may want to say "commit anyway".
Here's the example:

    class Sat a where
      dict :: a

    data EqD a = EqD {eq :: a->a->Bool}

    instance Sat (EqD a) => Eq a where
      (==) = eq dict

    instance Sat (EqD Integer) where
      dict = EqD{eq=(==)}

    instance Eq a => Sat (EqD a) where
      dict = EqD{eq=(==)}

    class Collection c cxt | c -> cxt where
      empty :: Sat (cxt a) => c a
      single :: Sat (cxt a) => a -> c a
      union :: Sat (cxt a) => c a -> c a -> c a
      member :: Sat (cxt a) => a -> c a -> Bool

    instance Collection [] EqD where
      empty = []
      single x = [x]
      union = (++)
      member = elem

It's an updated attempt to model "Restricted Data Types", if you
remember my Haskell workshop paper. In the end, though, GHC rejects
the program (even with fallow-overlapping-instances and
fallow-undecideable-instances), because there's more than one way to
construct the Eq instance needed by elem.

Yet all the ways are equivalent! So GHC is being a bit over-protective
of me, really: I know what I'm doing and I would LIKE it to pick an
arbitrary one. Maybe a flag fallow-incoherent-instances would be a
useful thing to add?

Correctly handle signed vs unsigned integer division on x86.

Documentation update to describe the new retainer profiling and
biographical profiling features.

Remove some temporary debugging code.

THIS CHANGE AFFECTS ALL OBJECT FILES.
Please say "make clean"

Move the entry code pointer to the start of info tables, according
to Plan C in ghc/rts/Storage.h

Fix various bugs in the implementation of subIntC and mulMayOflo.

Nuke -O2-for-C, since it can cause gcc to generate incorrect code, and
doesn't seem to help much.  Add a comment about this in the 5.02
release notes, saying you can still use -optc-O2 if you really want
to.

(merged from rev. 1.47.2.3)

merge rev. 1.11.4.1 from branch ghc-5-02-branch (remove -O2-for-C)

merge rev. 1.11.2.2 from ghc-5-02-branch (remove -O2-for-C).

remove unnecessary #include

Add the missing options list to the section "Replacing the program for
one or more phases".

Allow opening char & block devices with openFile.  Opening a char
device gets you a stream, and a block device a file (these are
heuristics since we can't really tell without trying to lseek() on the
device whether it supports seeking or not, but the defaults are at
least safe).

Redo translation of array indexing primops in order to insert 8 <-> 32
bit casts when dealing with Char arrays.  Also facilitate future removal
of CMem and MO_{Read,Write}OSBI.

In the process, notice and fix a large bogon in x86/sparc implementation
of signed and unsigned integer widening.

comment wibbles

Generate better code for case-of-literal (i.e. just do the
assignment).  These crop up now that the simplifier is a bit more
careful about duplicating literal strings.

mkLocalNonRec bug fix: we always want to generate a case when the rhs
of a binding is an unlifted type.  Previously the two cases for an
unlifted rhs and a strict binding were handled together which caused
some confusion, so now the cases are separate.

slightly clearer fix for previous bogon, and re-indent the function

Fix ifdefery bracketing bug on Win32

Re-added #ifdef which was lost in last commit