ghc-hetmet.git
23 years ago[project @ 2001-01-16 11:14:29 by simonmar]
simonmar [Tue, 16 Jan 2001 11:14:29 +0000 (11:14 +0000)]
[project @ 2001-01-16 11:14:29 by simonmar]
platform definitions for PPC/MacOSX

23 years ago[project @ 2001-01-16 10:51:39 by rrt]
rrt [Tue, 16 Jan 2001 10:51:39 +0000 (10:51 +0000)]
[project @ 2001-01-16 10:51:39 by rrt]
Fixed some typos.

23 years ago[project @ 2001-01-16 10:49:13 by simonmar]
simonmar [Tue, 16 Jan 2001 10:49:13 +0000 (10:49 +0000)]
[project @ 2001-01-16 10:49:13 by simonmar]
re-instate -Rghc-timing

23 years ago[project @ 2001-01-16 06:02:29 by qrczak]
qrczak [Tue, 16 Jan 2001 06:02:29 +0000 (06:02 +0000)]
[project @ 2001-01-16 06:02:29 by qrczak]
Oops, parens are bad in config.h values.

23 years ago[project @ 2001-01-15 22:24:54 by lewie]
lewie [Mon, 15 Jan 2001 22:24:54 +0000 (22:24 +0000)]
[project @ 2001-01-15 22:24:54 by lewie]
Just testing..

23 years ago[project @ 2001-01-15 20:55:14 by qrczak]
qrczak [Mon, 15 Jan 2001 20:55:15 +0000 (20:55 +0000)]
[project @ 2001-01-15 20:55:14 by qrczak]
Use -1 instead of 0 to mark unsupported errno names.
Add eOK for no error, i.e. corresponding to errno == 0.

23 years ago[project @ 2001-01-15 17:05:46 by sewardj]
sewardj [Mon, 15 Jan 2001 17:05:47 +0000 (17:05 +0000)]
[project @ 2001-01-15 17:05:46 by sewardj]
More stuff to do with primop support in the interpreter.  Also, track
some changes to the libraries.

23 years ago[project @ 2001-01-15 16:55:24 by sewardj]
sewardj [Mon, 15 Jan 2001 16:55:25 +0000 (16:55 +0000)]
[project @ 2001-01-15 16:55:24 by sewardj]
In interpreted code, basic support for routing primop calls through
to functions in PrelPrimopWrappers.lhs.

23 years ago[project @ 2001-01-15 09:57:33 by sewardj]
sewardj [Mon, 15 Jan 2001 09:57:33 +0000 (09:57 +0000)]
[project @ 2001-01-15 09:57:33 by sewardj]
Use mkApUpd0# to ensure top-level things are updateable.

23 years ago[project @ 2001-01-15 09:57:07 by sewardj]
sewardj [Mon, 15 Jan 2001 09:57:07 +0000 (09:57 +0000)]
[project @ 2001-01-15 09:57:07 by sewardj]
Handle nullary constructors more correctly.

23 years ago[project @ 2001-01-15 09:55:41 by sewardj]
sewardj [Mon, 15 Jan 2001 09:55:41 +0000 (09:55 +0000)]
[project @ 2001-01-15 09:55:41 by sewardj]
Add mkApUpd0# primop, used to make sure bytecode-compiled top-level things
are updateable.

23 years ago[project @ 2001-01-15 09:16:03 by simonmar]
simonmar [Mon, 15 Jan 2001 09:16:03 +0000 (09:16 +0000)]
[project @ 2001-01-15 09:16:03 by simonmar]
Improve the test for "still has jump to fast entry point": it was
turning up a false positive.

23 years ago[project @ 2001-01-15 07:33:02 by qrczak]
qrczak [Mon, 15 Jan 2001 07:33:02 +0000 (07:33 +0000)]
[project @ 2001-01-15 07:33:02 by qrczak]
Implemented #enum construct.

23 years ago[project @ 2001-01-14 15:46:53 by simonmar]
simonmar [Sun, 14 Jan 2001 15:46:53 +0000 (15:46 +0000)]
[project @ 2001-01-14 15:46:53 by simonmar]
Add a slightly stronger test for Time that the ones in tests/io.  This
one reads the current time, converts it to CalendarTime and back
again, and tests that all three times converted to strings compare equal.

23 years ago[project @ 2001-01-14 15:36:04 by simonmar]
simonmar [Sun, 14 Jan 2001 15:36:04 +0000 (15:36 +0000)]
[project @ 2001-01-14 15:36:04 by simonmar]
oops, fix a sign error that caused toClockTime to be wrong outside GMT.

23 years ago[project @ 2001-01-14 15:34:31 by simonmar]
simonmar [Sun, 14 Jan 2001 15:34:31 +0000 (15:34 +0000)]
[project @ 2001-01-14 15:34:31 by simonmar]
Patterns in the case statement didn't work on my FreeBSD box due to a
bogus-looking extra '(', so I removed it.  Hope I haven't broken anything.

23 years ago[project @ 2001-01-14 10:48:34 by panne]
panne [Sun, 14 Jan 2001 10:48:34 +0000 (10:48 +0000)]
[project @ 2001-01-14 10:48:34 by panne]
After Mr. HsLibs's std/lang library re-organization we need
hsc2hs-inplace in the "make boot" stage

23 years ago[project @ 2001-01-13 23:10:45 by qrczak]
qrczak [Sat, 13 Jan 2001 23:10:45 +0000 (23:10 +0000)]
[project @ 2001-01-13 23:10:45 by qrczak]
Don't output so many unnecessary C line markers.

23 years ago[project @ 2001-01-13 20:33:51 by qrczak]
qrczak [Sat, 13 Jan 2001 20:34:25 +0000 (20:34 +0000)]
[project @ 2001-01-13 20:33:51 by qrczak]
Use a replacement for LINE pragmas under nhc.

23 years ago[project @ 2001-01-13 19:46:49 by qrczak]
qrczak [Sat, 13 Jan 2001 19:46:49 +0000 (19:46 +0000)]
[project @ 2001-01-13 19:46:49 by qrczak]
Generate correct LINE pragmas.

23 years ago[project @ 2001-01-13 12:11:00 by qrczak]
qrczak [Sat, 13 Jan 2001 12:11:00 +0000 (12:11 +0000)]
[project @ 2001-01-13 12:11:00 by qrczak]
Remove #options construct: all released ghc versions understand multuple
{-# OPTIONS #-} pragmas.

23 years ago[project @ 2001-01-12 22:54:22 by qrczak]
qrczak [Fri, 12 Jan 2001 22:54:23 +0000 (22:54 +0000)]
[project @ 2001-01-12 22:54:22 by qrczak]
Expand #-constructs only outside Haskell comments and string literals.

23 years ago[project @ 2001-01-12 20:17:35 by qrczak]
qrczak [Fri, 12 Jan 2001 20:17:35 +0000 (20:17 +0000)]
[project @ 2001-01-12 20:17:35 by qrczak]
Define __GLASGOW_HASKELL__.

23 years ago[project @ 2001-01-12 18:49:25 by qrczak]
qrczak [Fri, 12 Jan 2001 18:49:25 +0000 (18:49 +0000)]
[project @ 2001-01-12 18:49:25 by qrczak]
Use $(CC) instead of ghc-inplace as C compiler in hsc2hs-inplace.
(Generally ghc is used here to obtain include dirs.)

23 years ago[project @ 2001-01-12 17:45:30 by qrczak]
qrczak [Fri, 12 Jan 2001 17:45:30 +0000 (17:45 +0000)]
[project @ 2001-01-12 17:45:30 by qrczak]
Add resetErrno :: IO ()
Use it in Directory, before calling readdir.

23 years ago[project @ 2001-01-12 17:38:27 by simonmar]
simonmar [Fri, 12 Jan 2001 17:38:27 +0000 (17:38 +0000)]
[project @ 2001-01-12 17:38:27 by simonmar]
need Integral CClock too.

23 years ago[project @ 2001-01-12 17:04:00 by simonmar]
simonmar [Fri, 12 Jan 2001 17:04:00 +0000 (17:04 +0000)]
[project @ 2001-01-12 17:04:00 by simonmar]
Add a comment about the tzname[] stuff

23 years ago[project @ 2001-01-12 16:44:13 by simonmar]
simonmar [Fri, 12 Jan 2001 16:44:13 +0000 (16:44 +0000)]
[project @ 2001-01-12 16:44:13 by simonmar]
re-order some of the conditional compilation & make it a bit more generic.

23 years ago[project @ 2001-01-12 16:40:07 by simonmar]
simonmar [Fri, 12 Jan 2001 16:40:07 +0000 (16:40 +0000)]
[project @ 2001-01-12 16:40:07 by simonmar]
fix toUTCTime, fill in some of the conditionally compiled bits

23 years ago[project @ 2001-01-12 16:16:36 by simonmar]
simonmar [Fri, 12 Jan 2001 16:16:37 +0000 (16:16 +0000)]
[project @ 2001-01-12 16:16:36 by simonmar]
Time library FFI'd.

23 years ago[project @ 2001-01-12 15:48:56 by simonmar]
simonmar [Fri, 12 Jan 2001 15:48:56 +0000 (15:48 +0000)]
[project @ 2001-01-12 15:48:56 by simonmar]
get CTime from PrelCTypesISO

23 years ago[project @ 2001-01-12 15:48:36 by simonmar]
simonmar [Fri, 12 Jan 2001 15:48:36 +0000 (15:48 +0000)]
[project @ 2001-01-12 15:48:36 by simonmar]
remove dup definition of CTime

23 years ago[project @ 2001-01-12 15:48:09 by simonmar]
simonmar [Fri, 12 Jan 2001 15:48:09 +0000 (15:48 +0000)]
[project @ 2001-01-12 15:48:09 by simonmar]
I need an Integral CTime.  Numeric only won't do.

23 years ago[project @ 2001-01-12 12:36:28 by simonmar]
simonmar [Fri, 12 Jan 2001 12:36:28 +0000 (12:36 +0000)]
[project @ 2001-01-12 12:36:28 by simonmar]
Add conversion tests to arith011.

23 years ago[project @ 2001-01-12 12:35:54 by simonmar]
simonmar [Fri, 12 Jan 2001 12:35:54 +0000 (12:35 +0000)]
[project @ 2001-01-12 12:35:54 by simonmar]
fix some bogus conversions.

23 years ago[project @ 2001-01-12 12:06:24 by sewardj]
sewardj [Fri, 12 Jan 2001 12:06:24 +0000 (12:06 +0000)]
[project @ 2001-01-12 12:06:24 by sewardj]
Latest bug fixes.

23 years ago[project @ 2001-01-12 12:04:53 by sewardj]
sewardj [Fri, 12 Jan 2001 12:04:53 +0000 (12:04 +0000)]
[project @ 2001-01-12 12:04:53 by sewardj]
Hopefully sort out heap-stack movement for constructors/cases.

23 years ago[project @ 2001-01-12 11:36:24 by simonpj]
simonpj [Fri, 12 Jan 2001 11:36:24 +0000 (11:36 +0000)]
[project @ 2001-01-12 11:36:24 by simonpj]
Amazingly, SimplCore.simplifyExpr didn't call the occurrence
analyser, so expressions typed in at the prompt were being
simplified very weedily.  Ditto the LHS and RHS of RULES.

So SimplCore.simplExprGently now does the business.  It runs
the simplifier twice because one pass leaves tons of crud.

This doesn't fix any bugs but it is cleaner.

23 years ago[project @ 2001-01-12 11:04:45 by simonmar]
simonmar [Fri, 12 Jan 2001 11:04:45 +0000 (11:04 +0000)]
[project @ 2001-01-12 11:04:45 by simonmar]
- Incorporate static STG flags (just -flet-no-escape at the mo) into
  hsc_minusO_flags and hsc_minusNoO_flags.

- Fix -v to be equivalent to -v3 (it was wrongly -v2 before).

- Make -v3 output hsc's list of static flags.

23 years ago[project @ 2001-01-12 10:18:14 by sewardj]
sewardj [Fri, 12 Jan 2001 10:18:14 +0000 (10:18 +0000)]
[project @ 2001-01-12 10:18:14 by sewardj]
Split ByteCodeGen up into more manageable-sized pieces.

23 years ago[project @ 2001-01-12 07:44:50 by qrczak]
qrczak [Fri, 12 Jan 2001 07:44:50 +0000 (07:44 +0000)]
[project @ 2001-01-12 07:44:50 by qrczak]
Adapt to the Addr/Ptr changes.
Throw away mkFastSubStringFO, mkFastSubStringFO#, eqStrPrefixFO.

23 years ago[project @ 2001-01-11 22:04:39 by qrczak]
qrczak [Thu, 11 Jan 2001 22:04:39 +0000 (22:04 +0000)]
[project @ 2001-01-11 22:04:39 by qrczak]
Allow Ptr in foreign dynamic declarations

23 years ago[project @ 2001-01-11 19:50:19 by qrczak]
qrczak [Thu, 11 Jan 2001 19:50:19 +0000 (19:50 +0000)]
[project @ 2001-01-11 19:50:19 by qrczak]
Remove backslash-newline pairs in #-constructs.

23 years ago[project @ 2001-01-11 19:48:28 by qrczak]
qrczak [Thu, 11 Jan 2001 19:48:28 +0000 (19:48 +0000)]
[project @ 2001-01-11 19:48:28 by qrczak]
CChar is Num so NUL can be written as 0.

23 years ago[project @ 2001-01-11 17:51:02 by simonmar]
simonmar [Thu, 11 Jan 2001 17:51:02 +0000 (17:51 +0000)]
[project @ 2001-01-11 17:51:02 by simonmar]
oops, I broke getArgs.  Unbreak it, and remove an unused import while
I'm here.

23 years ago[project @ 2001-01-11 17:35:24 by simonmar]
simonmar [Thu, 11 Jan 2001 17:35:24 +0000 (17:35 +0000)]
[project @ 2001-01-11 17:35:24 by simonmar]
add missing -u symbols.

23 years ago[project @ 2001-01-11 17:33:32 by simonmar]
simonmar [Thu, 11 Jan 2001 17:33:32 +0000 (17:33 +0000)]
[project @ 2001-01-11 17:33:32 by simonmar]
s/PrelAddr/Addr

23 years ago[project @ 2001-01-11 17:25:56 by simonmar]
simonmar [Thu, 11 Jan 2001 17:25:59 +0000 (17:25 +0000)]
[project @ 2001-01-11 17:25:56 by simonmar]
Re-organisation of ghc/lib/std and hslibs/lang
----------------------------------------------

In brief: move deprecated features out of ghc/lib/std and into
hslibs/lang, move new FFI libraries into ghc/lib/std and start
using them.

- foreign import may now return an unboxed type (this was
  advertised to work before, but in fact didn't).  Subsequent
  cleanups in PrelInt/PrelWord.

- Ptr is now defined in ghc/lib/std/PrelPtr.lhs.  Ptr is no
  longer a newtype of Addr, it is defined directly in terms of
  Addr#.

- PrelAddr has disappeared from ghc/lib/std, all uses of Addr in
  ghc/lib/std have been replaced with Ptr.  The definitions of
  Addr has been moved to hslibs/lang/Addr.lhs, as has
  lots of other Addr-related stuff.

- ForeignObj has been removed from ghc/lib/std, and replaced with
  ForeignPtr.  The definition of ForeignObj has been moved to
  hslibs/lang/ForeignObj.lhs.

- Most of the new FFI has been moved into ghc/lib/std in the form
  of modules PrelMarshalAlloc, PrelCString, PrelCError,
  PrelMarshalError, PrelMarshalArray, PrelMarshalUtils,
  PrelCTypes, PrelCTypesISO, and PrelStorable.  The corresponding
  modules in hslibs/lang simply re-export the contents of these
  modules.

- PrelPosixTypes defines a few POSIX types (CMode == mode_t,
  etc.)

- PrelCError changed to access errno using foreign label and peek
  (the POSIX book I have says that errno is guaranteed to be an
  extern int, so this should be OK until I get around to making
  errno thread-safe).

- Hacked the macros that generate the code for CTypes and
  CTypesISO to generate much less code
  (ghc/lib/std/cbits/CTypes.h).

- RtsAPI is now a bit more honest when it comes to building heap
  objects (it uses the correct constructors).

- the Bits class and related stuff has been moved to ghc/lib/std
  (it was simpler this way).

- Directory and System have been converted to use the new FFI.

23 years ago[project @ 2001-01-11 14:09:50 by simonpj]
simonpj [Thu, 11 Jan 2001 14:09:50 +0000 (14:09 +0000)]
[project @ 2001-01-11 14:09:50 by simonpj]
Add debug print for rule firings; controlled by -ddump-inlinings.

Also, make -ddump-inlinings work without -DDEBUG is off.
It's jolly useful, and it's tiresome to have to tell
people to rebuild their compiler.

23 years ago[project @ 2001-01-11 13:58:05 by simonpj]
simonpj [Thu, 11 Jan 2001 13:58:05 +0000 (13:58 +0000)]
[project @ 2001-01-11 13:58:05 by simonpj]
Fix a grotesque bug in the HsCore.pprUfExpr
The expression

((\x -> f x y) a b)

printed as

(\x -> f x y a b)

which in turn caused any importing module to
behave most oddly.

This didn't show up before, because such expressions don't
happen much (they are simplified).  But one showed up in a
RULE (which happens to be simplified only very gently), and
that tickled this bug.

23 years ago[project @ 2001-01-11 07:04:16 by qrczak]
qrczak [Thu, 11 Jan 2001 07:04:17 +0000 (07:04 +0000)]
[project @ 2001-01-11 07:04:16 by qrczak]
Change the representation of IOException: add optional filename.

23 years ago[project @ 2001-01-10 17:19:01 by sewardj]
sewardj [Wed, 10 Jan 2001 17:21:18 +0000 (17:21 +0000)]
[project @ 2001-01-10 17:19:01 by sewardj]
Today's interpreter bug fixes: FP stuff, and unpacking constrs onto stack.

23 years ago[project @ 2001-01-10 16:28:15 by qrczak]
qrczak [Wed, 10 Jan 2001 16:28:15 +0000 (16:28 +0000)]
[project @ 2001-01-10 16:28:15 by qrczak]
Remove CError.throwCError. Export the new function errnoToIOError.

Add a missing colon in a string inside IOException formed from
the error-specific part and optional filename. The IOException
representation has not changed yet, but IMHO the filename should
be kept separate.

Correctly handle user errors in IO.ioeGet* functions like other
IO errors.

Let fail :: String -> IO a throw a user error instead of invoking error
(Haskell98 is ambiguous about that and IMHO it should be this way;
this is also what nhc does).

23 years ago[project @ 2001-01-10 11:50:43 by rrt]
rrt [Wed, 10 Jan 2001 11:50:43 +0000 (11:50 +0000)]
[project @ 2001-01-10 11:50:43 by rrt]
Improved Windows installation instructions, thanks to a test drive by Andrew
Tolmach.

23 years ago[project @ 2001-01-09 17:43:57 by rrt]
rrt [Tue, 9 Jan 2001 17:43:57 +0000 (17:43 +0000)]
[project @ 2001-01-09 17:43:57 by rrt]
Add cMkDLL for --mk-dll switch.

23 years ago[project @ 2001-01-09 17:36:21 by sewardj]
sewardj [Tue, 9 Jan 2001 17:36:41 +0000 (17:36 +0000)]
[project @ 2001-01-09 17:36:21 by sewardj]
Various bug fixes for the interpreter/byte-code-gen combination.

23 years ago[project @ 2001-01-09 17:16:35 by rrt]
rrt [Tue, 9 Jan 2001 17:16:36 +0000 (17:16 +0000)]
[project @ 2001-01-09 17:16:35 by rrt]
Add --mk-dll.

23 years ago[project @ 2001-01-09 02:04:25 by chak]
chak [Tue, 9 Jan 2001 02:04:25 +0000 (02:04 +0000)]
[project @ 2001-01-09 02:04:25 by chak]
Due to problems with the pcl-cvs mode, I missed checking in these files, which
made my last commit break the build.  Sorry about that.

(Is anybody else using pcl-cvs v2.9.9 and getting strange parse errors when
updating from cvs.haskell.org?)

23 years ago[project @ 2001-01-08 12:58:45 by chak]
chak [Mon, 8 Jan 2001 12:58:45 +0000 (12:58 +0000)]
[project @ 2001-01-08 12:58:45 by chak]
Revision of Marcin's QError module

* I think, there was a rough consensus that Marcin's module provides the
  right kind of interface for handling "errno"s

* Malcolm preferred an enumeration, but Marcin made a convincing argument
  that this doesn't go well with architecture-dependent variations of "errno"

* Recoded the hsc magic into autoconf magic

* In hsc files, errno constants can be accessed directly; to provide access
  for ordinary Haskell programmer, constants for a range of common symbols are
  provided (their values are determined by configure)

* Introduced an `Errno' newtype to increase type safety

* renamed the throwIfXXX functions to throwErrnoIfXXX to have the original
  names available in the language independent MarshalError

* FIXME: We need a thread local getErrno in the RTS!

(It all compiles, but is still to be tested.)

23 years ago[project @ 2001-01-08 12:31:34 by rrt]
rrt [Mon, 8 Jan 2001 12:31:34 +0000 (12:31 +0000)]
[project @ 2001-01-08 12:31:34 by rrt]
Add -dynamic and make -static always the default.

23 years ago[project @ 2001-01-06 12:52:58 by panne]
panne [Sat, 6 Jan 2001 12:52:58 +0000 (12:52 +0000)]
[project @ 2001-01-06 12:52:58 by panne]
* Tab -> spaces
* Moved test for 4.11 to the correct place and revert Reuben's last fix.

23 years ago[project @ 2001-01-06 12:45:13 by panne]
panne [Sat, 6 Jan 2001 12:45:13 +0000 (12:45 +0000)]
[project @ 2001-01-06 12:45:13 by panne]
Make ghc/hslibs instead of happy the default projects to build/install
again.

23 years ago[project @ 2001-01-05 17:57:07 by rrt]
rrt [Fri, 5 Jan 2001 17:57:07 +0000 (17:57 +0000)]
[project @ 2001-01-05 17:57:07 by rrt]
Added some -Ks to keep 4.08 happy when compiling the compiler.

23 years ago[project @ 2001-01-05 17:56:23 by rrt]
rrt [Fri, 5 Jan 2001 17:56:23 +0000 (17:56 +0000)]
[project @ 2001-01-05 17:56:23 by rrt]
Windows fixes.

23 years ago[project @ 2001-01-05 16:06:01 by simonmar]
simonmar [Fri, 5 Jan 2001 16:06:01 +0000 (16:06 +0000)]
[project @ 2001-01-05 16:06:01 by simonmar]
Add some constraints to the tryRhsTyLam transformation to prevent it
floating out expressions of unlifted type.

23 years ago[project @ 2001-01-05 15:24:28 by sewardj]
sewardj [Fri, 5 Jan 2001 15:24:28 +0000 (15:24 +0000)]
[project @ 2001-01-05 15:24:28 by sewardj]
Various bug fixes.

23 years ago[project @ 2001-01-05 15:23:32 by sewardj]
sewardj [Fri, 5 Jan 2001 15:23:32 +0000 (15:23 +0000)]
[project @ 2001-01-05 15:23:32 by sewardj]
Various bug fixes, and implementation of string literals.

23 years ago[project @ 2001-01-04 19:43:07 by qrczak]
qrczak [Thu, 4 Jan 2001 19:43:07 +0000 (19:43 +0000)]
[project @ 2001-01-04 19:43:07 by qrczak]
Added --version parameter.
(QForeign must known whether to replace ghc's hsc2hs with its own.)

23 years ago[project @ 2001-01-04 16:22:55 by simonmar]
simonmar [Thu, 4 Jan 2001 16:22:55 +0000 (16:22 +0000)]
[project @ 2001-01-04 16:22:55 by simonmar]
add suffix rule for hsc2hs

23 years ago[project @ 2001-01-04 16:22:01 by simonmar]
simonmar [Thu, 4 Jan 2001 16:22:01 +0000 (16:22 +0000)]
[project @ 2001-01-04 16:22:01 by simonmar]
Add path to hsc2hs

23 years ago[project @ 2001-01-04 13:18:14 by simonmar]
simonmar [Thu, 4 Jan 2001 13:18:14 +0000 (13:18 +0000)]
[project @ 2001-01-04 13:18:14 by simonmar]
exit(1) in the case of a parse error.

23 years ago[project @ 2001-01-04 11:05:54 by qrczak]
qrczak [Thu, 4 Jan 2001 11:05:54 +0000 (11:05 +0000)]
[project @ 2001-01-04 11:05:54 by qrczak]
Fix bogus extra_libraries for rts (s/\[ "" \]/[]/).

23 years ago[project @ 2001-01-04 10:48:46 by simonmar]
simonmar [Thu, 4 Jan 2001 10:48:46 +0000 (10:48 +0000)]
[project @ 2001-01-04 10:48:46 by simonmar]
Fix the test for epilogue junk, and change it into a warning.

23 years ago[project @ 2001-01-04 08:42:21 by simonpj]
simonpj [Thu, 4 Jan 2001 08:42:21 +0000 (08:42 +0000)]
[project @ 2001-01-04 08:42:21 by simonpj]
Allow rules to apply for loop-breakers.
This backs out the previous (bogus) change.
There's a comment in Simplify.completeCall that explains.

23 years ago[project @ 2001-01-03 16:45:04 by sewardj]
sewardj [Wed, 3 Jan 2001 16:45:04 +0000 (16:45 +0000)]
[project @ 2001-01-03 16:45:04 by sewardj]
Updates to track bug fixes in the bytecode interpreter.

23 years ago[project @ 2001-01-03 16:44:29 by sewardj]
sewardj [Wed, 3 Jan 2001 16:44:30 +0000 (16:44 +0000)]
[project @ 2001-01-03 16:44:29 by sewardj]
Start getting the bytecode interpreter to work.  A matching commit to
compiler/ghci/ByteCodeGen.lhs follows ...

23 years ago[project @ 2001-01-03 16:04:16 by simonmar]
simonmar [Wed, 3 Jan 2001 16:04:16 +0000 (16:04 +0000)]
[project @ 2001-01-03 16:04:16 by simonmar]
Remove -ldl for now.

Current story: you don't get any GHCi stuff by default.  To build
GHCi:
  (a) build a compiler with GhcRtsHcOpts+=-optc-DGHCI in build.mk
  (b) use this compiler to bootstrap another compiler, with
      GhcWithInterpreter=YES and GhcHcOpts+=-ldl in build.mk

23 years ago[project @ 2001-01-03 15:39:32 by simonmar]
simonmar [Wed, 3 Jan 2001 15:39:32 +0000 (15:39 +0000)]
[project @ 2001-01-03 15:39:32 by simonmar]
Don't filter the contents of the package directories.  It was wrong
when the hi suffix was anything other than ".hi".

23 years ago[project @ 2001-01-03 15:30:48 by simonmar]
simonmar [Wed, 3 Jan 2001 15:30:48 +0000 (15:30 +0000)]
[project @ 2001-01-03 15:30:48 by simonmar]
make it compile with PROFILING turned on

23 years ago[project @ 2001-01-03 14:47:18 by simonmar]
simonmar [Wed, 3 Jan 2001 14:47:18 +0000 (14:47 +0000)]
[project @ 2001-01-03 14:47:18 by simonmar]
Correct more accidental abuse of the type system.

23 years ago[project @ 2001-01-03 14:38:08 by simonmar]
simonmar [Wed, 3 Jan 2001 14:38:08 +0000 (14:38 +0000)]
[project @ 2001-01-03 14:38:08 by simonmar]
When looking for an hi-boot file, look for "M.hi-boot" regardless of
which way we're compiling, rather than "M.<hi-suf>-boot" as it was
before.

Now the libs are hapily compiling for profiling.

23 years ago[project @ 2001-01-03 14:28:26 by simonmar]
simonmar [Wed, 3 Jan 2001 14:28:26 +0000 (14:28 +0000)]
[project @ 2001-01-03 14:28:26 by simonmar]
unbreak mkdependHS

23 years ago[project @ 2001-01-03 11:50:23 by simonpj]
simonpj [Wed, 3 Jan 2001 11:50:23 +0000 (11:50 +0000)]
[project @ 2001-01-03 11:50:23 by simonpj]
Add tc111

23 years ago[project @ 2001-01-03 11:48:06 by simonmar]
simonmar [Wed, 3 Jan 2001 11:48:06 +0000 (11:48 +0000)]
[project @ 2001-01-03 11:48:06 by simonmar]
don't try to preprocess non-haskellish source files.

23 years ago[project @ 2001-01-03 11:40:04 by simonpj]
simonpj [Wed, 3 Jan 2001 11:40:04 +0000 (11:40 +0000)]
[project @ 2001-01-03 11:40:04 by simonpj]
Dont quantify over monomorphic type variables

23 years ago[project @ 2001-01-03 11:32:33 by simonmar]
simonmar [Wed, 3 Jan 2001 11:32:33 +0000 (11:32 +0000)]
[project @ 2001-01-03 11:32:33 by simonmar]
Add a test for mixing unlifted types w/ polymorphic functions.

23 years ago[project @ 2001-01-03 11:18:51 by simonmar]
simonmar [Wed, 3 Jan 2001 11:18:53 +0000 (11:18 +0000)]
[project @ 2001-01-03 11:18:51 by simonmar]
s/boxed/lifted/

The typechecker's notion of "boxed" versus "unboxed" kind should
really have been "unlifted" versus "lifted" instead.  It is illegal to
unify an unlifted (but boxed) type with a polymorphic type variable,
since an unlifted/boxed type is always assumed to be a pointer to the
object itself, never a thunk or indirection.

This commit removes isUnboxedType, and renames a bunch of things that
were previously boxed/unboxed to unlifted/lifted.

23 years ago[project @ 2001-01-03 11:13:43 by simonmar]
simonmar [Wed, 3 Jan 2001 11:13:43 +0000 (11:13 +0000)]
[project @ 2001-01-03 11:13:43 by simonmar]
Hopefully fix the driver problems I introduced yesterday.

23 years ago[project @ 2001-01-03 10:57:41 by simonpj]
simonpj [Wed, 3 Jan 2001 10:57:41 +0000 (10:57 +0000)]
[project @ 2001-01-03 10:57:41 by simonpj]
Dont apply rules for a loop breaker

23 years ago[project @ 2001-01-03 08:34:32 by simonpj]
simonpj [Wed, 3 Jan 2001 08:34:32 +0000 (08:34 +0000)]
[project @ 2001-01-03 08:34:32 by simonpj]
Remove bogus zonkInst

23 years ago[project @ 2001-01-03 03:10:32 by chak]
chak [Wed, 3 Jan 2001 03:10:32 +0000 (03:10 +0000)]
[project @ 2001-01-03 03:10:32 by chak]
* Deprecated `Addr' and `freeHaskellFunctionPtr'

  I deprecated them only in the comments and docu.  Using a DEPRECATED pragma
  on the `Addr' functions doesn't make sense, because `Ptr' internally is
  build on top of `Addr'.

* Added `HsPtr' and `HsFunPtr' to `HsFFI.h'; deprecated `HsAddr'

* Updated the docu to reflect these changes

23 years ago[project @ 2001-01-02 16:42:10 by simonpj]
simonpj [Tue, 2 Jan 2001 16:42:10 +0000 (16:42 +0000)]
[project @ 2001-01-02 16:42:10 by simonpj]
Add a necessary isUnliftedType to completeBinding (coerce case)

23 years ago[project @ 2001-01-02 16:13:11 by simonmar]
simonmar [Tue, 2 Jan 2001 16:13:11 +0000 (16:13 +0000)]
[project @ 2001-01-02 16:13:11 by simonmar]
Remove redundant flag test

23 years ago[project @ 2001-01-02 15:30:57 by simonmar]
simonmar [Tue, 2 Jan 2001 15:30:57 +0000 (15:30 +0000)]
[project @ 2001-01-02 15:30:57 by simonmar]
Split the compilation pipeline into two stages, preprocessing and
compilation, so that OPTIONS pragmas which may affect the pipeline
work properly.  (now {-# OPTIONS -fvia-C #-} does the right thing in
one-shot mode as well as ghc --make mode).

23 years ago[project @ 2001-01-02 14:02:16 by simonmar]
simonmar [Tue, 2 Jan 2001 14:02:16 +0000 (14:02 +0000)]
[project @ 2001-01-02 14:02:16 by simonmar]
Be a bit stricter about what we treat as a #line pragma.

23 years ago[project @ 2001-01-02 12:30:51 by simonmar]
simonmar [Tue, 2 Jan 2001 12:31:25 +0000 (12:31 +0000)]
[project @ 2001-01-02 12:30:51 by simonmar]
revert last commit: don't import PprCore

23 years ago[project @ 2001-01-02 12:20:52 by simonmar]
simonmar [Tue, 2 Jan 2001 12:20:52 +0000 (12:20 +0000)]
[project @ 2001-01-02 12:20:52 by simonmar]
Update the comment next to the (disabled) epilogue junk test to
reflect the real reason this test should be disabled.

23 years ago[project @ 2001-01-02 11:20:35 by simonmar]
simonmar [Tue, 2 Jan 2001 11:20:35 +0000 (11:20 +0000)]
[project @ 2001-01-02 11:20:35 by simonmar]
GhcWithInterpreter defaults to NO for the time being.  (enable by hand
in build.mk to turn it on).

23 years ago[project @ 2000-12-31 16:58:05 by panne]
panne [Sun, 31 Dec 2000 16:58:05 +0000 (16:58 +0000)]
[project @ 2000-12-31 16:58:05 by panne]
Automagically add a GHC #define. These are the final changes to get a
bootstrapping GHC without a build.mk, thus enabling people (like me :-)
who forget obscure options quite often to build GHC painlessly.

Happy New Year!

23 years ago[project @ 2000-12-30 20:42:03 by panne]
panne [Sat, 30 Dec 2000 20:42:03 +0000 (20:42 +0000)]
[project @ 2000-12-30 20:42:03 by panne]
Don't install genprimopcode, it's not *that* interesting for the
average user...   :-)