simonmar [Thu, 18 Jan 2001 17:19:26 +0000 (17:19 +0000)]
[project @ 2001-01-18 17:19:26 by simonmar]
_scc_'s for the NCG
simonmar [Thu, 18 Jan 2001 16:55:03 +0000 (16:55 +0000)]
[project @ 2001-01-18 16:55:03 by simonmar]
add some _scc_'s
simonmar [Thu, 18 Jan 2001 16:30:01 +0000 (16:30 +0000)]
[project @ 2001-01-18 16:30:00 by simonmar]
Problem 1
=========
The typechecker, when deciding whether to extend the Package
environment with any new typechecked declarations in its hand, was
inserting new declarations into the environment only if the
declaration's module differed from the "current" module. This doesn't
work if the "current" module is a package module, as it could be at
the GHCi command line, for example.
The solution is to filter the declarations only if the current module
is not a package module.
Problem 2
=========
The "current" module, as obtained from the compilation manager, was
always bogusly a Home module (it used mkHomeModule). To properly fix
this, the GHCi state has to carry around Modules instead of
ModuleNames, and CompMan.cmLoadModule needs to return a list of
Modules.
simonmar [Thu, 18 Jan 2001 15:19:53 +0000 (15:19 +0000)]
[project @ 2001-01-18 15:19:42 by simonmar]
DocBook police
simonmar [Thu, 18 Jan 2001 15:19:22 +0000 (15:19 +0000)]
[project @ 2001-01-18 15:19:22 by simonmar]
update mailing-list info
simonmar [Thu, 18 Jan 2001 14:53:10 +0000 (14:53 +0000)]
[project @ 2001-01-18 14:52:07 by simonmar]
update mailing-list info
simonmar [Thu, 18 Jan 2001 14:50:52 +0000 (14:50 +0000)]
[project @ 2001-01-18 14:50:52 by simonmar]
Replace majordomo-specific instructions with mailman ones.
rrt [Thu, 18 Jan 2001 14:48:02 +0000 (14:48 +0000)]
[project @ 2001-01-18 14:48:02 by rrt]
Merge fix from ghc-4-07-branch (remove spurious bug-catching code that
messes up stderr files).
simonmar [Thu, 18 Jan 2001 12:54:17 +0000 (12:54 +0000)]
[project @ 2001-01-18 12:54:16 by simonmar]
Make the GHCi command line behave as if an "import qualified M" was in
force for all M.
The renamer now has a new "mode": CmdLineMode, which changes the
lookup machinery to turn a qualified lookup into an original name
lookup if the qualified name isn't otherwise in scope.
sewardj [Thu, 18 Jan 2001 12:27:42 +0000 (12:27 +0000)]
[project @ 2001-01-18 12:27:42 by sewardj]
Require happy-1.9 now.
sewardj [Thu, 18 Jan 2001 11:33:08 +0000 (11:33 +0000)]
[project @ 2001-01-18 11:33:08 by sewardj]
On sparc, do not make %i7 and %o7 available for allocation. Quite how
it ever worked before I don't know (and I bootstrapped the compiler with
the sparc NCG). Perhaps it was all just an illusion. Reminds me of how
Bobby Ewing was (utterly implausibly) bought back to life in Dallas.
sewardj [Thu, 18 Jan 2001 11:28:50 +0000 (11:28 +0000)]
[project @ 2001-01-18 11:28:50 by sewardj]
fix typo in comment
simonmar [Thu, 18 Jan 2001 11:16:08 +0000 (11:16 +0000)]
[project @ 2001-01-18 11:16:08 by simonmar]
When constructing a GlobalRdrEnv from an interface file for use from
the GHCi command line, don't export the qualified names: a module
which happens to export f from module M shouldn't bring into scope the
qualified identifier M.f when we use it from the command line.
simonmar [Thu, 18 Jan 2001 10:51:53 +0000 (10:51 +0000)]
[project @ 2001-01-18 10:51:53 by simonmar]
save the state after a ':t'
simonmar [Thu, 18 Jan 2001 10:31:38 +0000 (10:31 +0000)]
[project @ 2001-01-18 10:31:38 by simonmar]
add mkApUpd0zh_fast
simonmar [Thu, 18 Jan 2001 10:26:42 +0000 (10:26 +0000)]
[project @ 2001-01-18 10:26:42 by simonmar]
fix multiple instances of objects on the link line
sewardj [Wed, 17 Jan 2001 17:32:35 +0000 (17:32 +0000)]
[project @ 2001-01-17 17:32:35 by sewardj]
Update documentation for this magic program a little.
simonmar [Wed, 17 Jan 2001 16:54:04 +0000 (16:54 +0000)]
[project @ 2001-01-17 16:54:04 by simonmar]
Remove IfaceStuff and ParserStuff hacks, use happy-1.9's new multiple
%name feature. GHCi's command line isn't stuck in -fglasgow-exts mode
any more.
<CoverMyBack>
YOU NOW NEED HAPPY 1.9 TO BUILD GHC.
</CoverMyBack>
sewardj [Wed, 17 Jan 2001 16:46:02 +0000 (16:46 +0000)]
[project @ 2001-01-17 16:46:01 by sewardj]
Changes needed to get the head to compile on sparc-solaris.
sewardj [Wed, 17 Jan 2001 15:23:48 +0000 (15:23 +0000)]
[project @ 2001-01-17 15:23:39 by sewardj]
Bye bye STG Hugs!
simonmar [Wed, 17 Jan 2001 15:11:04 +0000 (15:11 +0000)]
[project @ 2001-01-17 15:11:04 by simonmar]
Update sweep
simonmar [Wed, 17 Jan 2001 12:46:13 +0000 (12:46 +0000)]
[project @ 2001-01-17 12:46:13 by simonmar]
dcs.gla.ac.uk ==> haskell.org
simonmar [Wed, 17 Jan 2001 12:14:30 +0000 (12:14 +0000)]
[project @ 2001-01-17 12:14:30 by simonmar]
merge changes from gmp-3.1.1
simonmar [Wed, 17 Jan 2001 12:06:21 +0000 (12:06 +0000)]
[project @ 2001-01-17 12:06:02 by simonmar]
Import stripped down gmp-3.1.1
simonmar [Wed, 17 Jan 2001 10:14:40 +0000 (10:14 +0000)]
[project @ 2001-01-17 10:14:40 by simonmar]
more wibbles
sewardj [Wed, 17 Jan 2001 10:11:17 +0000 (10:11 +0000)]
[project @ 2001-01-17 10:11:17 by sewardj]
Remove semicolon which was causing a syntax error on sparc-solaris /bin/sh.
simonmar [Wed, 17 Jan 2001 10:08:10 +0000 (10:08 +0000)]
[project @ 2001-01-17 10:08:10 by simonmar]
wibbles
simonmar [Wed, 17 Jan 2001 10:07:14 +0000 (10:07 +0000)]
[project @ 2001-01-17 10:07:14 by simonmar]
remove the message that says "now do gmake boot followed by gmake..."
qrczak [Tue, 16 Jan 2001 21:05:51 +0000 (21:05 +0000)]
[project @ 2001-01-16 21:05:51 by qrczak]
Correctly delimit a LINE pragma.
simonmar [Tue, 16 Jan 2001 17:47:10 +0000 (17:47 +0000)]
[project @ 2001-01-16 17:47:10 by simonmar]
Change the way the top-level Makefile works, now that we have a
dependency between ghc and hslibs which means that you can't `make
boot' in hslibs until you've done `make all' in ghc.
- now you just type `make all' at the top-level, and the makefile
arranges to do `make boot all' in each of the subdirectories.
Typing `make boot' at the top-level now yields a message explaining
what's going on.
- Now it's no longer necessary to set $(ProjectsToBuild). If you don't
set it, the build system attempts to build all the projects in
the source tree, in the order determined by $(AllProjects) in
config.mk. I've been meaning to fix this for ages, since it meant
that one had to hand-edit config.mk.in when making a distribution
for anything other than GHC - this is no longer the case.
You still *can* set ProjectsToBuild if you want to, however.
- ProjectsToInstall has been replaced by ProjectsDontInstall, and
has the obvious, reverse, meaning. It also has a reasonable default,
so the need to set it should be rare.
All this has the obvious benefit that to build GHC you need one fewer
commands (no more `make boot'), and anyone that has "./configure &&
make && make install" hardwired into their brains will feel right at
home.
sewardj [Tue, 16 Jan 2001 17:09:43 +0000 (17:09 +0000)]
[project @ 2001-01-16 17:09:43 by sewardj]
Various ghci interactive UI fixes/improvements.
sewardj [Tue, 16 Jan 2001 15:44:42 +0000 (15:44 +0000)]
[project @ 2001-01-16 15:44:42 by sewardj]
hscExpr: be a bit more careful about which pcs version is returned, so
that the environment isn't totally hosed following a type error.
simonmar [Tue, 16 Jan 2001 14:16:44 +0000 (14:16 +0000)]
[project @ 2001-01-16 14:16:44 by simonmar]
Improve a parser error message.
simonmar [Tue, 16 Jan 2001 14:06:14 +0000 (14:06 +0000)]
[project @ 2001-01-16 14:06:14 by simonmar]
Fix a problem with our use of tcsetattr() which caused background
processes to stop when the buffering mode, or the echo status, was
changed.
tcSetAttr() is a wrapper around tcsetattr() that temporarily blocks
SIGTTOU around the call.
sewardj [Tue, 16 Jan 2001 12:44:34 +0000 (12:44 +0000)]
[project @ 2001-01-16 12:44:34 by sewardj]
Give the correct commit message for rev 1.59:
"rm debug printing"
sewardj [Tue, 16 Jan 2001 12:42:19 +0000 (12:42 +0000)]
[project @ 2001-01-16 12:42:18 by sewardj]
Fill in some more missing cases.
sewardj [Tue, 16 Jan 2001 12:41:05 +0000 (12:41 +0000)]
[project @ 2001-01-16 12:41:05 by sewardj]
rm
simonmar [Tue, 16 Jan 2001 12:41:03 +0000 (12:41 +0000)]
[project @ 2001-01-16 12:41:03 by simonmar]
take account of GhcUnregisterised
simonmar [Tue, 16 Jan 2001 12:32:45 +0000 (12:32 +0000)]
[project @ 2001-01-16 12:32:45 by simonmar]
add GhcUnregisterised to Config.hs
simonmar [Tue, 16 Jan 2001 12:31:18 +0000 (12:31 +0000)]
[project @ 2001-01-16 12:31:18 by simonmar]
Add GhcUnregisterised, to make unregisterised compilation the default.
Also include a lengthy comment about how the whole unregisterised
compilation / boostrapping thing is supposed to hold together.
simonmar [Tue, 16 Jan 2001 12:02:04 +0000 (12:02 +0000)]
[project @ 2001-01-16 12:02:04 by simonmar]
Don't use the same name for a typedef and variable.
simonmar [Tue, 16 Jan 2001 11:59:06 +0000 (11:59 +0000)]
[project @ 2001-01-16 11:59:06 by simonmar]
this isn't legal C, apparently:
switch (e) {
label:
}
without at least one statement between the label and the end block.
simonmar [Tue, 16 Jan 2001 11:57:06 +0000 (11:57 +0000)]
[project @ 2001-01-16 11:57:06 by simonmar]
make the dummy Callable and Returnable static arrays one element long,
because some C compilers can't cope with zero-length static arrays.
simonmar [Tue, 16 Jan 2001 11:54:25 +0000 (11:54 +0000)]
[project @ 2001-01-16 11:54:25 by simonmar]
add MaxOSX version of mmap.
simonmar [Tue, 16 Jan 2001 11:53:32 +0000 (11:53 +0000)]
[project @ 2001-01-16 11:53:32 by simonmar]
add MacOSX HEAP_BASE
simonmar [Tue, 16 Jan 2001 11:50:30 +0000 (11:50 +0000)]
[project @ 2001-01-16 11:50:30 by simonmar]
Don't use the same name for a typedef and variable. Some C compilers
can handle this without complaint (eg. gcc), but some others it seems
can't.
simonmar [Tue, 16 Jan 2001 11:28:45 +0000 (11:28 +0000)]
[project @ 2001-01-16 11:28:45 by simonmar]
add dummy freeHaskellFunctionPtr when we don't have a real one on this arch.
simonmar [Tue, 16 Jan 2001 11:27:45 +0000 (11:27 +0000)]
[project @ 2001-01-16 11:27:45 by simonmar]
temp. bug fix for MacOS X which has a large off_t. This code is
broken anyway.
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
rrt [Tue, 16 Jan 2001 10:51:39 +0000 (10:51 +0000)]
[project @ 2001-01-16 10:51:39 by rrt]
Fixed some typos.
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
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.
lewie [Mon, 15 Jan 2001 22:24:54 +0000 (22:24 +0000)]
[project @ 2001-01-15 22:24:54 by lewie]
Just testing..
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.
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.
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.
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.
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.
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.
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.
qrczak [Mon, 15 Jan 2001 07:33:02 +0000 (07:33 +0000)]
[project @ 2001-01-15 07:33:02 by qrczak]
Implemented #enum construct.
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.
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.
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.
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
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.
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.
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.
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.
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.
qrczak [Fri, 12 Jan 2001 20:17:35 +0000 (20:17 +0000)]
[project @ 2001-01-12 20:17:35 by qrczak]
Define __GLASGOW_HASKELL__.
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.)
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.
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.
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
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.
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
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.
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
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
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.
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.
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.
sewardj [Fri, 12 Jan 2001 12:06:24 +0000 (12:06 +0000)]
[project @ 2001-01-12 12:06:24 by sewardj]
Latest bug fixes.
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.
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.
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.
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.
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.
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
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.
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.
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.
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.
simonmar [Thu, 11 Jan 2001 17:33:32 +0000 (17:33 +0000)]
[project @ 2001-01-11 17:33:32 by simonmar]
s/PrelAddr/Addr
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.
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.
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.
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.
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.