ghc-hetmet.git
22 years ago[project @ 2001-07-24 10:37:57 by simonmar]
simonmar [Tue, 24 Jul 2001 10:37:57 +0000 (10:37 +0000)]
[project @ 2001-07-24 10:37:57 by simonmar]
Remove whitespace from the end of lines.  The extra whitespace is
syntactically meaningful to make(!) and was causing us to do
a superfluous mkdependC in directories without any C files.

22 years ago[project @ 2001-07-24 10:31:07 by simonmar]
simonmar [Tue, 24 Jul 2001 10:31:07 +0000 (10:31 +0000)]
[project @ 2001-07-24 10:31:07 by simonmar]
libm is a dependency of the RTS now, so don't also depend on it for
package "std".

22 years ago[project @ 2001-07-24 10:28:51 by simonmar]
simonmar [Tue, 24 Jul 2001 10:28:51 +0000 (10:28 +0000)]
[project @ 2001-07-24 10:28:51 by simonmar]
back the last change out; a better way to get all the hc files you need is

SRC_HC_OPTS += -keep-hc-files

in your build.mk.

22 years ago[project @ 2001-07-24 09:55:16 by simonpj]
simonpj [Tue, 24 Jul 2001 09:55:16 +0000 (09:55 +0000)]
[project @ 2001-07-24 09:55:16 by simonpj]
Two fixes to the demand analyis.

1.  Don't look inside Coerces.  These wrap recursive newtypes,
and we might fail to find a fixpoint if we look inside them.

2.  Polymorphic 'seq' is represented by a (Seq _ _ []) demand,
so lub/both must be prepared to take a Seq with an empty list
of demands without complaint.

22 years ago[project @ 2001-07-24 09:53:27 by simonpj]
simonpj [Tue, 24 Jul 2001 09:53:27 +0000 (09:53 +0000)]
[project @ 2001-07-24 09:53:27 by simonpj]
Keep the printed stuff for Demand and DmdResult separate, else
we can get confused when parsing a strictness signature.  For
example, previously

LX

could have been parsed as L  -> X
or as LX -> T

Urk!  I've taken the simple path of making DmdResults print in
lower case.  Thus

Lx  means L  -> x
LX means LX -> t

22 years ago[project @ 2001-07-24 09:46:55 by simonmar]
simonmar [Tue, 24 Jul 2001 09:46:55 +0000 (09:46 +0000)]
[project @ 2001-07-24 09:46:55 by simonmar]
Add Ken Shan to the contributor's list for his heroic work on the
Alpha port of GHC.  Thanks Ken!

22 years ago[project @ 2001-07-24 09:08:47 by simonpj]
simonpj [Tue, 24 Jul 2001 09:08:47 +0000 (09:08 +0000)]
[project @ 2001-07-24 09:08:47 by simonpj]
Add missing cases to newDemand

22 years ago[project @ 2001-07-24 06:31:35 by ken]
ken [Tue, 24 Jul 2001 06:31:36 +0000 (06:31 +0000)]
[project @ 2001-07-24 06:31:35 by ken]
Innocent changes to resurrect/add 64-bit support.

22 years ago[project @ 2001-07-24 06:02:21 by ken]
ken [Tue, 24 Jul 2001 06:02:21 +0000 (06:02 +0000)]
[project @ 2001-07-24 06:02:21 by ken]
Fix typo in StgRun() for the Alpha

22 years ago[project @ 2001-07-24 06:01:21 by ken]
ken [Tue, 24 Jul 2001 06:01:21 +0000 (06:01 +0000)]
[project @ 2001-07-24 06:01:21 by ken]
#include <string.h> to get proper prototypes

22 years ago[project @ 2001-07-24 05:53:27 by ken]
ken [Tue, 24 Jul 2001 05:53:27 +0000 (05:53 +0000)]
[project @ 2001-07-24 05:53:27 by ken]
Removed debugging (tracing) code.

22 years ago[project @ 2001-07-24 05:49:32 by ken]
ken [Tue, 24 Jul 2001 05:49:32 +0000 (05:49 +0000)]
[project @ 2001-07-24 05:49:32 by ken]
Fixed punctuation typo.

22 years ago[project @ 2001-07-24 05:17:52 by ken]
ken [Tue, 24 Jul 2001 05:17:52 +0000 (05:17 +0000)]
[project @ 2001-07-24 05:17:52 by ken]
Added -I$(GHC_INCLUDE_DIR) to $(SRC_MKDEPENDC_OPTS), so that
mkdependC can find its include files properly.

22 years ago[project @ 2001-07-24 05:08:51 by ken]
ken [Tue, 24 Jul 2001 05:08:51 +0000 (05:08 +0000)]
[project @ 2001-07-24 05:08:51 by ken]
Made "nat" 64-bit ("unsigned long") rather than 32-bit ("unsigned int")
on 64-bit machines.  This fixes some problems caused by code elsewhere
assuming that sizeof(nat) == sizeof(void *).

22 years ago[project @ 2001-07-24 05:07:37 by ken]
ken [Tue, 24 Jul 2001 05:07:37 +0000 (05:07 +0000)]
[project @ 2001-07-24 05:07:37 by ken]
Added -lm to the extra_libraries list for the rts package.  This is
because ldexp() is used by StgPrimFloat.c and is in -lm on alpha-osf3.

22 years ago[project @ 2001-07-24 05:05:41 by ken]
ken [Tue, 24 Jul 2001 05:05:41 +0000 (05:05 +0000)]
[project @ 2001-07-24 05:05:41 by ken]
Added $(GhcHcOpts) to $(SRC_HC_OPTS), so that by setting
--keep-hc-files-too in $(GhcHcOpts) we get .hc files for the
Haskell code under this directory, ghc/driver.

22 years ago[project @ 2001-07-24 05:04:58 by ken]
ken [Tue, 24 Jul 2001 05:04:59 +0000 (05:04 +0000)]
[project @ 2001-07-24 05:04:58 by ken]
Removed 32-bit dependencies in the generation and handling of
liveness mask bitmaps.  We now support both 32-bit and 64-bit
machines with identical .hc files.  Support for >64-bit machines
would be easy to add.  Note that old .hc files are incompatible
with the changes made to ghc/include/InfoMacros.h!

22 years ago[project @ 2001-07-24 04:52:49 by ken]
ken [Tue, 24 Jul 2001 04:52:49 +0000 (04:52 +0000)]
[project @ 2001-07-24 04:52:49 by ken]
Made some code #ifdef GHCI, so that it will compile without GHCI enabled.

22 years ago[project @ 2001-07-24 04:47:06 by ken]
ken [Tue, 24 Jul 2001 04:47:06 +0000 (04:47 +0000)]
[project @ 2001-07-24 04:47:06 by ken]
Added missing hyphen in frnot of machdepCCOpts for mips and powerpc.

22 years ago[project @ 2001-07-24 04:46:37 by ken]
ken [Tue, 24 Jul 2001 04:46:37 +0000 (04:46 +0000)]
[project @ 2001-07-24 04:46:37 by ken]
Added -Xlinker -noprefix_recognition to Alpha-specific
machdepCCOpts.  We need this flag to disable the following
documented behavior of ld(1) on Digital UNIX:

   All routines that start with __init_ are treated, by default,
   as initialization routines. These are routines that are called
   without an argument when the file that contains them is loaded
   or when the program that contains them is started.

Unfortunately, this flag produces the annoying (but harmless) gcc
message

    gcc: -noprefix_recognition: linker input file unused since
    linking not done

(See also mk/bootstrap.mk)

22 years ago[project @ 2001-07-24 04:45:59 by ken]
ken [Tue, 24 Jul 2001 04:45:59 +0000 (04:45 +0000)]
[project @ 2001-07-24 04:45:59 by ken]
Added -static, a flag for compiling on the Alpha that is already
listed under machdepCCOpts in DriverFlags.hs.  Without it, the
following assert in GC.c fails:

    /* make sure the info pointer is into text space */
    ASSERT(q && (LOOKS_LIKE_GHC_INFO(GET_INFO(q))
                 || IS_HUGS_CONSTR_INFO(GET_INFO(q))));

Added -Xlinker -noprefix_recognition, a flag for Alpha compilation
that I just added to machdepCCOpts in DriverFlags.hs.  We need
this flag to disable the following documented behavior of ld(1) on
Digital UNIX:

   All routines that start with __init_ are treated, by default,
   as initialization routines. These are routines that are called
   without an argument when the file that contains them is loaded
   or when the program that contains them is started.

Unfortunately, this flag produces the annoying (but harmless) gcc
message

    gcc: -noprefix_recognition: linker input file unused since
    linking not done

(See also ghc/compiler/main/DriverFlags.hs)

22 years ago[project @ 2001-07-24 04:41:40 by ken]
ken [Tue, 24 Jul 2001 04:41:40 +0000 (04:41 +0000)]
[project @ 2001-07-24 04:41:40 by ken]
Added #include "SchedAPI.h" to fix compiler warning message.

22 years ago[project @ 2001-07-24 04:39:31 by ken]
ken [Tue, 24 Jul 2001 04:39:32 +0000 (04:39 +0000)]
[project @ 2001-07-24 04:39:31 by ken]
Make the Time module thread-safe by calling the reentrant functions
gmtime_r and localtime_r instead of gmtime and localtime wherever
they are available.

(This is necessary to make Time work at all on our Alpha machine --
perhaps GHC tickles the reentrancy of the C library or something?)

22 years ago[project @ 2001-07-24 04:35:36 by ken]
ken [Tue, 24 Jul 2001 04:35:36 +0000 (04:35 +0000)]
[project @ 2001-07-24 04:35:36 by ken]
Fixed: The type of sec and usec in struct timeval are always 32 bits,
but not always a CLong.  (On Alphas, CLong is 64 bits.)

22 years ago[project @ 2001-07-23 23:37:35 by andy]
andy [Mon, 23 Jul 2001 23:37:35 +0000 (23:37 +0000)]
[project @ 2001-07-23 23:37:35 by andy]
Adding the -Pa flag. This allows you to see all the cost centres

22 years ago[project @ 2001-07-23 23:29:47 by ken]
ken [Mon, 23 Jul 2001 23:29:47 +0000 (23:29 +0000)]
[project @ 2001-07-23 23:29:47 by ken]
Removed 32-bit assumptions.

22 years ago[project @ 2001-07-23 23:27:50 by ken]
ken [Mon, 23 Jul 2001 23:27:50 +0000 (23:27 +0000)]
[project @ 2001-07-23 23:27:50 by ken]
- Added recommended gcc warning flag: -Wcast-align

  - Fixed: "make distclean" would run in the gmp subdirectory multiple
    times, once for each way.  The second time, the Makefile in there
    is gone, causing an error.

22 years ago[project @ 2001-07-23 23:26:14 by ken]
ken [Mon, 23 Jul 2001 23:26:14 +0000 (23:26 +0000)]
[project @ 2001-07-23 23:26:14 by ken]
Changed comment to reflect reality: We now use 32-bit thread IDs, not
64-bit ones.

22 years ago[project @ 2001-07-23 23:14:58 by ken]
ken [Mon, 23 Jul 2001 23:14:58 +0000 (23:14 +0000)]
[project @ 2001-07-23 23:14:58 by ken]
Removed 32-bit assumptions in info table macros (mostly replacing
StgWord32 with StgWord).

22 years ago[project @ 2001-07-23 23:13:07 by ken]
ken [Mon, 23 Jul 2001 23:13:07 +0000 (23:13 +0000)]
[project @ 2001-07-23 23:13:07 by ken]
Brought the mangler up-to-date for the Alpha.

22 years ago[project @ 2001-07-23 23:10:03 by ken]
ken [Mon, 23 Jul 2001 23:10:03 +0000 (23:10 +0000)]
[project @ 2001-07-23 23:10:03 by ken]
Fixed: The second-stage compilation in this script should use the compiler
built in the first stage.  Also, configure wants to find Happy, so we feed
it fake-happy (presuming that the ghc-less machine also does not have happy).

22 years ago[project @ 2001-07-23 23:08:41 by sof]
sof [Mon, 23 Jul 2001 23:08:41 +0000 (23:08 +0000)]
[project @ 2001-07-23 23:08:41 by sof]
lex_demand: added a couple of missing cases for new demand type

22 years ago[project @ 2001-07-23 23:08:04 by ken]
ken [Mon, 23 Jul 2001 23:08:04 +0000 (23:08 +0000)]
[project @ 2001-07-23 23:08:04 by ken]
Rearranged the usage help message to move --no-compile to the bottom.

More importantly-- Fixed: hsc2hs was translating

    #def inline ... function(...) { ... }

into (assuming defined(__GNUC__))

    [in the _hsc.h file:]  extern inline ... function(...)
    [in the _hsc.c file:]  inline ... function(...)

while it should really be translated into

    [in the _hsc.h file:]  extern inline ... function(...)
    [in the _hsc.c file:]  ... function(...)

(without the "inline" in the _hsc.c file).  Quoting the gcc documentation
for "inline":

>    If you specify both `inline' and `extern' in the function
> definition, then the definition is used only for inlining.  In no case
> is the function compiled on its own, not even if you refer to its
> address explicitly.  Such an address becomes an external reference, as
> if you had only declared the function, and had not defined it.
>
>    This combination of `inline' and `extern' has almost the effect of a
> macro.  The way to use it is to put a function definition in a header
> file with these keywords, and put another copy of the definition
> (lacking `inline' and `extern') in a library file.  The definition in
> the header file will cause most calls to the function to be inlined.
> If any uses of the function remain, they will refer to the single copy
> in the library.

22 years ago[project @ 2001-07-23 23:02:37 by ken]
ken [Mon, 23 Jul 2001 23:02:37 +0000 (23:02 +0000)]
[project @ 2001-07-23 23:02:37 by ken]
Added documentation for hsc2hs "--no-compile" option.

22 years ago[project @ 2001-07-23 22:56:44 by ken]
ken [Mon, 23 Jul 2001 22:56:44 +0000 (22:56 +0000)]
[project @ 2001-07-23 22:56:44 by ken]
Add the *.a.list "list of object files to archive" file
to MOSTLY_CLEAN_FILES.

22 years ago[project @ 2001-07-23 22:52:33 by ken]
ken [Mon, 23 Jul 2001 22:52:33 +0000 (22:52 +0000)]
[project @ 2001-07-23 22:52:33 by ken]
Improved how we build library archives:

On Digital UNIX (or where otherwise available), use the "Z" option to
compress archives as we build them.

On Digital UNIX 5 (or where otherwise available), use the "-input" option,
which reads from a specified file the list of objects to archive.  This is
significantly faster than piping the list of objects to xargs when the
number of objects to archive is large.

22 years ago[project @ 2001-07-23 22:38:02 by ken]
ken [Mon, 23 Jul 2001 22:38:02 +0000 (22:38 +0000)]
[project @ 2001-07-23 22:38:02 by ken]
Fixed: making dependencies for Haskell source files fails if -split-objs
is present in $(HC_OPTS).

22 years ago[project @ 2001-07-23 22:33:52 by ken]
ken [Mon, 23 Jul 2001 22:33:53 +0000 (22:33 +0000)]
[project @ 2001-07-23 22:33:52 by ken]
Added "--enable-hc-boot-unregisterised" option to configure,
for bootstrapping from unregisterised HC files.

22 years ago[project @ 2001-07-23 22:26:18 by ken]
ken [Mon, 23 Jul 2001 22:26:18 +0000 (22:26 +0000)]
[project @ 2001-07-23 22:26:18 by ken]
Teach configure.in to canonicalise new-fangled Alpha system names,
such as alphaev67-dec-osf5.0, to alpha-dec-osf3.

22 years ago[project @ 2001-07-23 22:22:58 by ken]
ken [Mon, 23 Jul 2001 22:22:58 +0000 (22:22 +0000)]
[project @ 2001-07-23 22:22:58 by ken]
Added a helper script "fake-happy" that fakes "happy -v" output.
It is useful for bootstrapping on new platforms.

22 years ago[project @ 2001-07-23 22:19:59 by ken]
ken [Mon, 23 Jul 2001 22:19:59 +0000 (22:19 +0000)]
[project @ 2001-07-23 22:19:59 by ken]
Grab the latest versions of config.guess and config.sub from GNU.

22 years ago[project @ 2001-07-23 21:31:00 by sof]
sof [Mon, 23 Jul 2001 21:31:00 +0000 (21:31 +0000)]
[project @ 2001-07-23 21:31:00 by sof]
ranlib gmp/libgmp.a - needed for a.out plats (OpenBSD in particular).

22 years ago[project @ 2001-07-23 20:19:53 by sof]
sof [Mon, 23 Jul 2001 20:19:53 +0000 (20:19 +0000)]
[project @ 2001-07-23 20:19:53 by sof]
Handle leading underscores for PrelMain undefineds

22 years ago[project @ 2001-07-23 18:25:48 by sof]
sof [Mon, 23 Jul 2001 18:25:48 +0000 (18:25 +0000)]
[project @ 2001-07-23 18:25:48 by sof]
Get rid off DEBUG - prev commit didn't attack the real problem

22 years ago[project @ 2001-07-23 17:28:33 by simonmar]
simonmar [Mon, 23 Jul 2001 17:28:33 +0000 (17:28 +0000)]
[project @ 2001-07-23 17:28:33 by simonmar]
Back out change that tried to sneak in with the new garbage collector.

22 years ago[project @ 2001-07-23 17:23:19 by simonmar]
simonmar [Mon, 23 Jul 2001 17:23:20 +0000 (17:23 +0000)]
[project @ 2001-07-23 17:23:19 by simonmar]
Add a compacting garbage collector.

It isn't enabled by default, as there are still a couple of problems:
there's a fallback case I haven't implemented yet which means it will
occasionally bomb out, and speed-wise it's quite a bit slower than the
copying collector (about 1.8x slower).

Until I can make it go faster, it'll only be useful when you're
actually running low on real memory.

'+RTS -c' to enable it.

Oh, and I cleaned up a few things in the RTS while I was there, and
fixed one or two possibly real bugs in the existing GC.

22 years ago[project @ 2001-07-23 16:27:41 by simonpj]
simonpj [Mon, 23 Jul 2001 16:27:41 +0000 (16:27 +0000)]
[project @ 2001-07-23 16:27:41 by simonpj]
Wibble

22 years ago[project @ 2001-07-23 16:16:47 by sof]
sof [Mon, 23 Jul 2001 16:16:47 +0000 (16:16 +0000)]
[project @ 2001-07-23 16:16:47 by sof]
Mismatched module name in non-DEBUG case

22 years ago[project @ 2001-07-23 15:11:55 by simonmar]
simonmar [Mon, 23 Jul 2001 15:11:55 +0000 (15:11 +0000)]
[project @ 2001-07-23 15:11:55 by simonmar]
Grrr.  Do not #include "Stg.h" in plain C code, because that enables
all the global register variables.  #include "Rts.h" directly instead.

This was correct in the original StgLongLong.c, why on earth it was
changed is beyond me.

22 years ago[project @ 2001-07-23 13:49:49 by simonmar]
simonmar [Mon, 23 Jul 2001 13:49:49 +0000 (13:49 +0000)]
[project @ 2001-07-23 13:49:49 by simonmar]
add a couple of missing ';'

22 years ago[project @ 2001-07-23 11:44:15 by qrczak]
qrczak [Mon, 23 Jul 2001 11:44:15 +0000 (11:44 +0000)]
[project @ 2001-07-23 11:44:15 by qrczak]
import qualified Demand

22 years ago[project @ 2001-07-23 10:54:46 by simonpj]
simonpj [Mon, 23 Jul 2001 10:54:50 +0000 (10:54 +0000)]
[project @ 2001-07-23 10:54:46 by simonpj]
---------------------------------
Switch to the new demand analyser
---------------------------------

This commit makes the new demand analyser the main beast,
with the old strictness analyser as a backup.  When
DEBUG is on, the old strictness analyser is run too, and the
results compared.

WARNING: this isn't thorougly tested yet, so expect glitches.
Delay updating for a few days if the HEAD is mission critical
for you.

But do try it out.  I'm away for 2.5 weeks from Thursday, so
it would be good to shake out any glaring bugs before then.

22 years ago[project @ 2001-07-23 10:48:56 by simonpj]
simonpj [Mon, 23 Jul 2001 10:48:56 +0000 (10:48 +0000)]
[project @ 2001-07-23 10:48:56 by simonpj]
block/unblockAsynchExns have arity 2

22 years ago[project @ 2001-07-23 10:47:16 by simonmar]
simonmar [Mon, 23 Jul 2001 10:47:16 +0000 (10:47 +0000)]
[project @ 2001-07-23 10:47:16 by simonmar]
Small changes to improve GC performance slightly:

  - store the generation *number* in the block descriptor rather
    than a pointer to the generation structure, since the most
    common operation is to pull out the generation number, and
    it's one less indirection this way.

  - cache the generation number in the step structure too, which
    avoids an extra indirection in several places.

22 years ago[project @ 2001-07-23 10:43:42 by simonmar]
simonmar [Mon, 23 Jul 2001 10:43:42 +0000 (10:43 +0000)]
[project @ 2001-07-23 10:43:42 by simonmar]
The real version of this file is in ghc/includes/.

22 years ago[project @ 2001-07-23 10:43:18 by simonmar]
simonmar [Mon, 23 Jul 2001 10:43:18 +0000 (10:43 +0000)]
[project @ 2001-07-23 10:43:18 by simonmar]
merge in changes from rts/Linker.h

22 years ago[project @ 2001-07-23 10:42:37 by simonmar]
simonmar [Mon, 23 Jul 2001 10:42:37 +0000 (10:42 +0000)]
[project @ 2001-07-23 10:42:37 by simonmar]
Small fixes to the help text for the -i heap profiling interval option.

22 years ago[project @ 2001-07-23 10:24:57 by simonpj]
simonpj [Mon, 23 Jul 2001 10:24:58 +0000 (10:24 +0000)]
[project @ 2001-07-23 10:24:57 by simonpj]
Yet another newtype-squashing bug; this time TcType.unifyTyX

22 years ago[project @ 2001-07-22 03:28:25 by chak]
chak [Sun, 22 Jul 2001 03:28:25 +0000 (03:28 +0000)]
[project @ 2001-07-22 03:28:25 by chak]
ELF_TARGET_386 has to be #defined for x86/Solaris.

22 years ago[project @ 2001-07-22 03:19:51 by chak]
chak [Sun, 22 Jul 2001 03:19:51 +0000 (03:19 +0000)]
[project @ 2001-07-22 03:19:51 by chak]
Solaris system headers contain function definitions (not just prototypes) for
some functions; eg, stat().  WDP decided to toss (some of) them out during
mangling.  I think, this is wrong, as the list of these functions changes with
varying Solaris releases and it makes foreign importing these functions more
complicated; in particular, it broke Directory.hsc and PosixFiles on
x86/Solaris.

22 years ago[project @ 2001-07-21 19:08:44 by sof]
sof [Sat, 21 Jul 2001 19:08:44 +0000 (19:08 +0000)]
[project @ 2001-07-21 19:08:44 by sof]
ghc-inplace.bat: COMMAND.COM doesn't support %*, fall back to using %[1-9]

22 years ago[project @ 2001-07-21 11:38:19 by simonmar]
simonmar [Sat, 21 Jul 2001 11:38:19 +0000 (11:38 +0000)]
[project @ 2001-07-21 11:38:19 by simonmar]
make ==> $(MAKE)

22 years ago[project @ 2001-07-20 16:48:20 by simonpj]
simonpj [Fri, 20 Jul 2001 16:48:21 +0000 (16:48 +0000)]
[project @ 2001-07-20 16:48:20 by simonpj]
This commit adds the very convenient function

  Subst.substTyWith :: [TyVar] -> [Type] -> Type -> Type

and uses it in various places.

22 years ago[project @ 2001-07-20 16:47:55 by simonpj]
simonpj [Fri, 20 Jul 2001 16:47:55 +0000 (16:47 +0000)]
[project @ 2001-07-20 16:47:55 by simonpj]
------------------------
More newtype squashing
------------------------

Recursive newtypes were confusing the worker/wrapper generator.
This is because I originally got rid of opaque newtypes altogether,
then put them back for recursive ones only, and forgot to reinstate
the cunning stuff in the w/w stuff.

(Discovered by Sigbjorn; thanks!)

22 years ago[project @ 2001-07-20 15:22:21 by simonpj]
simonpj [Fri, 20 Jul 2001 15:22:22 +0000 (15:22 +0000)]
[project @ 2001-07-20 15:22:21 by simonpj]
-----------------------
Get rid of ArityAtLeast
-----------------------

Now that we have CgInfo, with the exact code-generator arity
for the value, we don't need the distinction between ArityAtLeast
and ArityExactly in the ArityInfo field of an IdInfo.

This commit makes

type ArityInfo = Maybe Arity

and propagates this change consistently through the compiler.

22 years ago[project @ 2001-07-20 10:09:32 by simonpj]
simonpj [Fri, 20 Jul 2001 10:09:32 +0000 (10:09 +0000)]
[project @ 2001-07-20 10:09:32 by simonpj]
Third cut at the demand analyser; seems to work nicely now

22 years ago[project @ 2001-07-20 10:08:56 by simonpj]
simonpj [Fri, 20 Jul 2001 10:08:56 +0000 (10:08 +0000)]
[project @ 2001-07-20 10:08:56 by simonpj]
--------------------------------------------
Hack around an infelicity in the simplifier
-------------------------------------------

Use 3 iterations of the simplifier (instead of the previous 2)
because otherwise we land up with huge dead expressions.
Reason: an infelicity in the simpifier.

      let k = BIG in foldr k z xs
 ==>  let k = BIG in letrec go = \xs -> ...(k x).... in go xs
 ==>  let k = BIG in letrec go = \xs -> ...(BIG x).... in go xs
 Don't stop now!

I couldn't see an easy way to make the simplifier avoid this
big intermediate, so I hacked round it like this.

22 years ago[project @ 2001-07-20 10:05:37 by simonpj]
simonpj [Fri, 20 Jul 2001 10:05:37 +0000 (10:05 +0000)]
[project @ 2001-07-20 10:05:37 by simonpj]
--------------------------
Correct a bug in exprArity
--------------------------

This long-standing bug meant that exprArity gave over-pessimistic
answers.  e.g.

\x -> f x

where nothing is known about f.  We were getting

0 (for f)
-1 (for the argument)
+1 (for the lambda)
-----
0

The right answer is of course 1.

22 years ago[project @ 2001-07-20 01:05:11 by sof]
sof [Fri, 20 Jul 2001 01:05:11 +0000 (01:05 +0000)]
[project @ 2001-07-20 01:05:11 by sof]
lookupSymbol (PEi386-only): When chasing down a label in a DLL, try
removing the initial underscore (if any). I've prefixed the comment
next to the change with the label 'HACK', as I'm not sure whether
this is a name mangling operation you should always do when going
from the import library name to the DLL-bound thing.
(Julian, could you have a look?)

Fixes probs with 'foreign import'ing DLL-bound symbols.

22 years ago[project @ 2001-07-20 00:44:33 by sof]
sof [Fri, 20 Jul 2001 00:44:33 +0000 (00:44 +0000)]
[project @ 2001-07-20 00:44:33 by sof]
Since longlong prims are no longer in the RTS, extend RTS_MINGW_ONLY_SYMBOLS with some reqd libgcc syms

22 years ago[project @ 2001-07-19 22:08:07 by sof]
sof [Thu, 19 Jul 2001 22:08:07 +0000 (22:08 +0000)]
[project @ 2001-07-19 22:08:07 by sof]
More FPTOOLS_END_{TEXT,DATA}_SECTION fixes

22 years ago[project @ 2001-07-19 20:27:23 by sof]
sof [Thu, 19 Jul 2001 20:27:23 +0000 (20:27 +0000)]
[project @ 2001-07-19 20:27:23 by sof]
FPTOOLS_END_{TEXT,DATA}_SECTION: fix 'test -n' usage

22 years ago[project @ 2001-07-19 15:34:39 by apt]
apt [Thu, 19 Jul 2001 15:34:39 +0000 (15:34 +0000)]
[project @ 2001-07-19 15:34:39 by apt]
fix typo in double2IntLit

22 years ago[project @ 2001-07-19 15:33:17 by apt]
apt [Thu, 19 Jul 2001 15:33:17 +0000 (15:33 +0000)]
[project @ 2001-07-19 15:33:17 by apt]
external core: omit repn for recursive newtypes and fix char literals

22 years ago[project @ 2001-07-19 15:32:05 by apt]
apt [Thu, 19 Jul 2001 15:32:05 +0000 (15:32 +0000)]
[project @ 2001-07-19 15:32:05 by apt]
reinstate inlinings that are completely within an INLINE

22 years ago[project @ 2001-07-19 14:23:16 by rrt]
rrt [Thu, 19 Jul 2001 14:23:16 +0000 (14:23 +0000)]
[project @ 2001-07-19 14:23:16 by rrt]
More fixes to allow GHC to live in C:\Program Files

22 years ago[project @ 2001-07-19 09:26:33 by simonpj]
simonpj [Thu, 19 Jul 2001 09:26:33 +0000 (09:26 +0000)]
[project @ 2001-07-19 09:26:33 by simonpj]
Next cut at the demand analyser

22 years ago[project @ 2001-07-19 09:26:08 by simonpj]
simonpj [Thu, 19 Jul 2001 09:26:09 +0000 (09:26 +0000)]
[project @ 2001-07-19 09:26:08 by simonpj]
Move eqListBy to Util, and use it

22 years ago[project @ 2001-07-19 09:24:56 by simonpj]
simonpj [Thu, 19 Jul 2001 09:24:56 +0000 (09:24 +0000)]
[project @ 2001-07-19 09:24:56 by simonpj]
Output a slightly better message for -ddump-hi-diffs

22 years ago[project @ 2001-07-19 09:03:24 by simonmar]
simonmar [Thu, 19 Jul 2001 09:03:24 +0000 (09:03 +0000)]
[project @ 2001-07-19 09:03:24 by simonmar]
Fix a number of bugs in the autoconf setup, thanks to Gregory Wright
<gwright@packetstorm.com>.

22 years ago[project @ 2001-07-19 09:01:40 by simonmar]
simonmar [Thu, 19 Jul 2001 09:01:40 +0000 (09:01 +0000)]
[project @ 2001-07-19 09:01:40 by simonmar]
remove a vestige of the long-gone FLOATS_AS_DOUBLES

22 years ago[project @ 2001-07-19 08:28:59 by andy]
andy [Thu, 19 Jul 2001 08:28:59 +0000 (08:28 +0000)]
[project @ 2001-07-19 08:28:59 by andy]
import of hscThing needs to be conditional on GHCi status.

22 years ago[project @ 2001-07-19 07:28:00 by andy]
andy [Thu, 19 Jul 2001 07:28:00 +0000 (07:28 +0000)]
[project @ 2001-07-19 07:28:00 by andy]
Re-adding the -i flag for heap profiling.

The problem was that profiling forces 2-space collection,
and the census was done on *every* GC. Now, we still keep
the 2-space for profiling, but just do the census periodically,
according to a new flag, -i<secs>.

22 years ago[project @ 2001-07-18 19:32:15 by sof]
sof [Wed, 18 Jul 2001 19:32:15 +0000 (19:32 +0000)]
[project @ 2001-07-18 19:32:15 by sof]
Restrict ghc-4.xx breakage to mingw only; not an issue for other plats

22 years ago[project @ 2001-07-18 16:06:10 by rrt]
rrt [Wed, 18 Jul 2001 16:06:11 +0000 (16:06 +0000)]
[project @ 2001-07-18 16:06:10 by rrt]
Add support for Hugs's :info command. Doesn't work yet, but shouldn't
interfere with anything else. Some of the files touched are just to correct
out-of-date comments.

Highlights are:

hscThing: like hscStmt, but just gets info about a single identifier
cmInfoThing: exposes hscThing's functionality to the outside world

22 years ago[project @ 2001-07-18 16:04:31 by rrt]
rrt [Wed, 18 Jul 2001 16:04:31 +0000 (16:04 +0000)]
[project @ 2001-07-18 16:04:31 by rrt]
Fix typo (variable name wrong; wouldn't build)

22 years ago[project @ 2001-07-17 23:02:14 by sof]
sof [Tue, 17 Jul 2001 23:02:14 +0000 (23:02 +0000)]
[project @ 2001-07-17 23:02:14 by sof]
wibble - the joys of CVS conflict merging

22 years ago[project @ 2001-07-17 21:12:33 by qrczak]
qrczak [Tue, 17 Jul 2001 21:12:33 +0000 (21:12 +0000)]
[project @ 2001-07-17 21:12:33 by qrczak]
Include way in HSC_C_OBJS to clean *_hsc.p_o too.

22 years ago[project @ 2001-07-17 18:47:52 by sof]
sof [Tue, 17 Jul 2001 18:47:52 +0000 (18:47 +0000)]
[project @ 2001-07-17 18:47:52 by sof]
wibble

22 years ago[project @ 2001-07-17 17:27:20 by sof]
sof [Tue, 17 Jul 2001 17:27:21 +0000 (17:27 +0000)]
[project @ 2001-07-17 17:27:20 by sof]
Drop the dependency on the Win32 library, as it is only used in an
extremely limited way (a single 'foreign import' is all that's reqd)
[Sorry Reuben, I don't buy your 'extra deps are good' argument :-) ]

22 years ago[project @ 2001-07-17 17:12:04 by sof]
sof [Tue, 17 Jul 2001 17:12:04 +0000 (17:12 +0000)]
[project @ 2001-07-17 17:12:04 by sof]
runSomething: narrow down prev commit, only quote args under mingw;it interacts badly with -D quoting already done on other plats (and isn't reqd)

22 years ago[project @ 2001-07-17 15:48:40 by simonpj]
simonpj [Tue, 17 Jul 2001 15:48:40 +0000 (15:48 +0000)]
[project @ 2001-07-17 15:48:40 by simonpj]
Add comment and use backslash consistently

22 years ago[project @ 2001-07-17 15:48:21 by simonpj]
simonpj [Tue, 17 Jul 2001 15:48:21 +0000 (15:48 +0000)]
[project @ 2001-07-17 15:48:21 by simonpj]
Remove trace

22 years ago[project @ 2001-07-17 15:28:30 by simonpj]
simonpj [Tue, 17 Jul 2001 15:28:31 +0000 (15:28 +0000)]
[project @ 2001-07-17 15:28:30 by simonpj]
--------------------------------
First cut at the demand analyser
--------------------------------

This demand analyser is intended to replace the strictness/absence
analyser, and the CPR analyser.

This commit adds it to the compiler, but in an entirely non-invasive
way.

If you build the compiler without -DDEBUG,
you won't get it at all.

If you build the compiler with -DDEBUG,
you'll get the demand analyser, but the existing
strictness analyser etc are still there.  All the
demand analyser does is to compare its output with
the existing stuff and report differences.

There's no cross-module stuff for demand info yet.

The strictness/demand info is put the IdInfo as
newStrictnessInfo
newDemandInfo

Eventually we'll remove the old ones.

Simon

22 years ago[project @ 2001-07-17 14:53:48 by rrt]
rrt [Tue, 17 Jul 2001 14:53:48 +0000 (14:53 +0000)]
[project @ 2001-07-17 14:53:48 by rrt]
Reset the terminal on exit so that ghci doesn't leave it messed up. To see
what I mean, run ghci, then type :q RET, then su return, and start typing a
password. Alarming, isn't it?

This fix needs a patch to the Readline library (next commit).

22 years ago[project @ 2001-07-17 14:51:53 by rrt]
rrt [Tue, 17 Jul 2001 14:51:53 +0000 (14:51 +0000)]
[project @ 2001-07-17 14:51:53 by rrt]
Make -u's and symbols into separate args

22 years ago[project @ 2001-07-17 14:48:04 by rrt]
rrt [Tue, 17 Jul 2001 14:48:04 +0000 (14:48 +0000)]
[project @ 2001-07-17 14:48:04 by rrt]
1. Make GHC work with paths and filenames containing spaces (hopefully).
This is done by quoting all arguments sent to the shell. To avoid causing
problems, this means that each argument in a list had *better* be a single
argument, hence the next commit to PackageSrc.hs.

2. undosify the path passed as the -h argument to unlit, to make unlit work
on Windows again.

3. Fix getExecDir for Windows (broken by previous changes to the meaning of
-B).

22 years ago[project @ 2001-07-17 09:55:09 by qrczak]
qrczak [Tue, 17 Jul 2001 09:55:09 +0000 (09:55 +0000)]
[project @ 2001-07-17 09:55:09 by qrczak]
Typos in a comment. Whitespace at eols.

22 years ago[project @ 2001-07-17 09:36:55 by qrczak]
qrczak [Tue, 17 Jul 2001 09:36:55 +0000 (09:36 +0000)]
[project @ 2001-07-17 09:36:55 by qrczak]
Fix comment.

22 years ago[project @ 2001-07-17 03:53:25 by qrczak]
qrczak [Tue, 17 Jul 2001 03:53:25 +0000 (03:53 +0000)]
[project @ 2001-07-17 03:53:25 by qrczak]
Long long functions are no longer in the rts.