simonmar [Wed, 25 Jul 2001 12:18:26 +0000 (12:18 +0000)]
[project @ 2001-07-25 12:18:26 by simonmar]
- move the call to scavenge_mark_stack() inside the inner scavenge loop,
so it gets called more often and there's less chance of the mark
stack filling up. None of the nofib tests cause the mark stack to
fill up now.
- remove some experimental code from scavenge_mark_stack().
simonmar [Wed, 25 Jul 2001 11:55:57 +0000 (11:55 +0000)]
[project @ 2001-07-25 11:55:57 by simonmar]
Specialise thread_static() a little bit: improves performance for
threading the static object list.
simonpj [Wed, 25 Jul 2001 10:49:46 +0000 (10:49 +0000)]
[project @ 2001-07-25 10:49:46 by simonpj]
Do for evalStrictness the same as for evalAbsence
simonmar [Wed, 25 Jul 2001 10:10:25 +0000 (10:10 +0000)]
[project @ 2001-07-25 10:10:25 by simonmar]
- Move FptoolsHcOpts out of the hslibs section of this file (which was
*exactly* the wrong place for it).
- Clarify the meaning of FptoolsHcOpts and GhcHcOpts.
NOTE: $(FptoolsHcOpts) gets added to every Haskell compilation for
Haskell *programs* (not libraries), including GHC itself. By default,
it contains only -O. To remove the -O, you need to override
FptoolsHcOpts in build.mk, not GhcHcOpts.
GhcHcOpts is still there, and contains options to be added when
compiling GHC only.
simonpj [Wed, 25 Jul 2001 09:19:41 +0000 (09:19 +0000)]
[project @ 2001-07-25 09:19:41 by simonpj]
---------------------------------
Another long-standing infelicity!
---------------------------------
CoreTidy was throwing away demand info on let-binders.
This meant that CorePrep would build a let (thunk) instead of
a case, even if the Id is sure to be demanded.
Easily fixed by making CoreTidy retain the demand info.
This demand-analysis stuff is having the excellent side effect
of flushing out performance bugs!
simonmar [Wed, 25 Jul 2001 09:14:21 +0000 (09:14 +0000)]
[project @ 2001-07-25 09:14:21 by simonmar]
- bugfix (was erroneously ignoring the return value from scavenge_one() and
checking failed_to_evac instead, which was always false)
- printf format fix (shut gcc up).
simonpj [Wed, 25 Jul 2001 08:04:00 +0000 (08:04 +0000)]
[project @ 2001-07-25 08:04:00 by simonpj]
Change pretty-print of DmdType slightly
simonpj [Wed, 25 Jul 2001 07:43:53 +0000 (07:43 +0000)]
[project @ 2001-07-25 07:43:53 by simonpj]
---------------------------
Fix another bad DmdAnal bug
---------------------------
The `both` operator wasn't commutative, leading to most
strange results. In particular, the fixpoint finder went into
an infinite fip/flop loop on Marcin's program.
simonpj [Wed, 25 Jul 2001 07:42:23 +0000 (07:42 +0000)]
[project @ 2001-07-25 07:42:23 by simonpj]
---------------------------------------
Fix a grevious dict-inlining infelicity
---------------------------------------
This commit fixes an infelicity that must have been there
for a long time. The problem was that dictionary functions
(arising from instance declarations) were treated as GlobalIds,
and hence no taken account of when doing free variable and dependency
analysis. As a result, the specialiser was messing up dependency
order, so the program was considerably more complex than it should
be by the time it got to the strictness analyser.
Net result:
PrelRead.$sreduce :: Integer -> Integer -> (Integer,Integer)
didn't get detected as strict.
Easily fixed, by making DictFunIds into LocalIds (albeit exported ones).
sof [Tue, 24 Jul 2001 18:04:01 +0000 (18:04 +0000)]
[project @ 2001-07-24 18:04:01 by sof]
Capture the dependency of PrelIO.hsc on PrelHandle_hsc.h (as was, 'make clean; make all' broke)
simonpj [Tue, 24 Jul 2001 16:47:40 +0000 (16:47 +0000)]
[project @ 2001-07-24 16:47:40 by simonpj]
Print a bit more info in an old strictness analyser assert fail
simonpj [Tue, 24 Jul 2001 16:46:51 +0000 (16:46 +0000)]
[project @ 2001-07-24 16:46:51 by simonpj]
Print a bit more info in the dmdFix loop
simonpj [Tue, 24 Jul 2001 16:42:11 +0000 (16:42 +0000)]
[project @ 2001-07-24 16:42:11 by simonpj]
A major demand-analyser fix, which made it say something was
absent when it wasn't at all. Here's the comment from the
Case equation of dmdAnal.
-- Figure out whether the case binder is used, and use
-- that to set the keepity of the demand. This is utterly essential.
-- Consider f x = case x of y { (a,b) -> k y a }
-- If we just take scrut_demand = U(L,A), then we won't pass x to the
-- worker, so the worker will rebuild
-- x = (a, absent-error)
-- and that'll crash.
simonmar [Tue, 24 Jul 2001 16:36:44 +0000 (16:36 +0000)]
[project @ 2001-07-24 16:36:43 by simonmar]
Bugfixes; take large objects into account in stats output.
simonpj [Tue, 24 Jul 2001 16:09:48 +0000 (16:09 +0000)]
[project @ 2001-07-24 16:09:48 by simonpj]
wibble
simonpj [Tue, 24 Jul 2001 16:04:21 +0000 (16:04 +0000)]
[project @ 2001-07-24 16:04:21 by simonpj]
Add errorCString
simonpj [Tue, 24 Jul 2001 15:57:27 +0000 (15:57 +0000)]
[project @ 2001-07-24 15:57:27 by simonpj]
Make absent-arg errors more descriptive
simonmar [Tue, 24 Jul 2001 15:13:01 +0000 (15:13 +0000)]
[project @ 2001-07-24 15:13:01 by simonmar]
More tweaks. Getting usable now.
simonmar [Tue, 24 Jul 2001 14:29:13 +0000 (14:29 +0000)]
[project @ 2001-07-24 14:29:13 by simonmar]
Make compacting GC faster, stage 1.
A cunning trick from Simon P.J. to avoid needing to call the
potentially expensive LOOKS_LIKE_GHC_INFO() a lot.
simonmar [Tue, 24 Jul 2001 10:38:48 +0000 (10:38 +0000)]
[project @ 2001-07-24 10:38:48 by simonmar]
remove extra SRC_MKDEPENDC_OPTS, since we shouldn't be doing a
mkdependC in this directory at all.
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.
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".
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.
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.
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
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!
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
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.
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
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
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.
ken [Tue, 24 Jul 2001 05:49:32 +0000 (05:49 +0000)]
[project @ 2001-07-24 05:49:32 by ken]
Fixed punctuation typo.
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.
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 *).
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.
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.
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!
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.
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.
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)
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)
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.
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?)
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.)
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
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.
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.
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.
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).
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.
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).
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
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.
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.
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.
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.
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).
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.
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.
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.
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.
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).
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
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
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.
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.
simonpj [Mon, 23 Jul 2001 16:27:41 +0000 (16:27 +0000)]
[project @ 2001-07-23 16:27:41 by simonpj]
Wibble
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
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.
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 ';'
qrczak [Mon, 23 Jul 2001 11:44:15 +0000 (11:44 +0000)]
[project @ 2001-07-23 11:44:15 by qrczak]
import qualified Demand
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.
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
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.
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/.
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
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.
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
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.
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.
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]
simonmar [Sat, 21 Jul 2001 11:38:19 +0000 (11:38 +0000)]
[project @ 2001-07-21 11:38:19 by simonmar]
make ==> $(MAKE)
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.
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!)
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.
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
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.
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.
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.
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
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
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
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
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
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
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
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
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
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
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>.