Thomas Schilling [Mon, 13 Oct 2008 17:06:58 +0000 (17:06 +0000)]
Use cabal information to get GHC's flags to `ghctags'.
By giving the dist-directory to ghctags we can get all the GHC API
flags we need in order to load the required modules. The flag name
could perhaps be improved, but apart from that it seems to work well.
Clemens Fruhwirth [Tue, 14 Oct 2008 08:13:00 +0000 (08:13 +0000)]
Version bump for libffi to 3.0.6
Clemens Fruhwirth [Mon, 13 Oct 2008 22:15:30 +0000 (22:15 +0000)]
Encode shared/static configuration into stamps to do the right thing when rebuilding
Ian Lynagh [Mon, 13 Oct 2008 19:59:43 +0000 (19:59 +0000)]
Add a link to the GHC API docs from the library haddock index
Ian Lynagh [Mon, 13 Oct 2008 20:09:27 +0000 (20:09 +0000)]
Link to the GHC API documentation from the main doc page
Ian Lynagh [Mon, 13 Oct 2008 20:06:25 +0000 (20:06 +0000)]
Whitespace only in docs/index.html
Ian Lynagh [Mon, 13 Oct 2008 19:25:48 +0000 (19:25 +0000)]
Tweak gen_contents_index
It now works again after it has been installed, as well as while it is
in a source tree.
After it's been installed it filters out the ghc package, as that
currently swamps everything else in the index.
Clemens Fruhwirth [Mon, 13 Oct 2008 20:16:08 +0000 (20:16 +0000)]
Build fixes for DLLized rts
Clemens Fruhwirth [Mon, 13 Oct 2008 20:14:26 +0000 (20:14 +0000)]
Do not filter the rts from linked libraries in linkDynLib as Windows does not allow unresolved symbols
Clemens Fruhwirth [Mon, 13 Oct 2008 20:09:45 +0000 (20:09 +0000)]
Add HsFFI.o to INSTALL_LIBS
Clemens Fruhwirth [Mon, 13 Oct 2008 16:24:33 +0000 (16:24 +0000)]
Rename symbol macros to a consistant naming scheme
Simon Marlow [Mon, 13 Oct 2008 12:13:39 +0000 (12:13 +0000)]
Fix #2685: two Bool arguments to tidyTypeEnv were the wrong way around
So -XTemplateHaskell was behaving like -fomit-interface-file-pragmas,
and vice versa.
Ian Lynagh [Sat, 11 Oct 2008 19:10:08 +0000 (19:10 +0000)]
Simplify the "is $bindir in $PATH" test
Ian Lynagh [Sat, 11 Oct 2008 19:10:30 +0000 (19:10 +0000)]
Correct the "is $bindir in $PATH" test
We were testing neq instead of eq
pepe [Sat, 11 Oct 2008 11:47:20 +0000 (11:47 +0000)]
Fix a typo which was causing ghci to quit on commands errors
Clemens Fruhwirth [Sat, 11 Oct 2008 07:45:24 +0000 (07:45 +0000)]
Drop libm from the linker dependencies for libffi
Clemens Fruhwirth [Fri, 10 Oct 2008 19:23:18 +0000 (19:23 +0000)]
Do not generate haddock documentation when running install-docs in libffi
Josef Svenningsson [Fri, 10 Oct 2008 15:03:22 +0000 (15:03 +0000)]
When waking up thread blocked on TVars, wake oldest first (#2319)
StgTVarWatchQueue contains the threads blocked on a TVar in order
youngest first. The list has to be traversed backwards to unpark the threads
oldest first.
This improves the fairness when using STM in some situations.
Simon Marlow [Fri, 10 Oct 2008 13:15:45 +0000 (13:15 +0000)]
add readTVarIO :: TVar a -> IO a
Simon Marlow [Fri, 10 Oct 2008 13:15:35 +0000 (13:15 +0000)]
fix #2636: throw missing module errors as SourceErrors, not ErrMsg
Simon Marlow [Wed, 8 Oct 2008 15:47:02 +0000 (15:47 +0000)]
atomicModifyIORef: use a local cas() instead of the global lock
This should improve scaling when using atomicModifyIORef
Clemens Fruhwirth [Fri, 10 Oct 2008 07:31:06 +0000 (07:31 +0000)]
Delay building libffi until package.conf is created and fix bindist
Ian Lynagh [Thu, 9 Oct 2008 16:49:46 +0000 (16:49 +0000)]
Install a versioned ghc-pkg script; fixes trac #2662
Ian Lynagh [Thu, 9 Oct 2008 16:34:51 +0000 (16:34 +0000)]
Fix bindist creation: Only the main RTS was being put in the bindists
Simon Marlow [Thu, 9 Oct 2008 09:11:18 +0000 (09:11 +0000)]
pushAtom: add missing case for MachNullAddr (#2589)
Simon Marlow [Thu, 9 Oct 2008 08:51:18 +0000 (08:51 +0000)]
undo incorrect assertion, and fix comments
Simon Marlow [Thu, 9 Oct 2008 08:50:51 +0000 (08:50 +0000)]
remove old GRAN/PARALLEL_HASKELL code
Manuel M T Chakravarty [Thu, 9 Oct 2008 13:23:28 +0000 (13:23 +0000)]
FIX #2639
MERGE TO 6.10
Manuel M T Chakravarty [Thu, 9 Oct 2008 06:14:35 +0000 (06:14 +0000)]
Cover PredTy case in Type.tyFamInsts
MERGE TO 6.10
Clemens Fruhwirth [Thu, 9 Oct 2008 07:13:42 +0000 (07:13 +0000)]
Drop ghcconfig.h/RtsConfig.h from libffi's package.conf.in
Clemens Fruhwirth [Wed, 8 Oct 2008 23:44:55 +0000 (23:44 +0000)]
Don't use sed's -i flag as Solaris doesn't know it in libffi/Makefile
Clemens Fruhwirth [Wed, 8 Oct 2008 23:29:02 +0000 (23:29 +0000)]
Don't use /dev/null trick to create empty object files in libffi/Makefile
Clemens Fruhwirth [Wed, 8 Oct 2008 17:04:43 +0000 (17:04 +0000)]
Turn libffi into a Haskell package
Thomas Schilling [Wed, 8 Oct 2008 14:40:32 +0000 (14:40 +0000)]
Make 'getModSummary' deterministic.
Thomas Schilling [Tue, 7 Oct 2008 23:56:56 +0000 (23:56 +0000)]
Add accessors to 'HsModule' and haddockify it.
Simon Marlow [Wed, 8 Oct 2008 10:34:32 +0000 (10:34 +0000)]
fix syntax errors in src-dist publish rules
Simon Marlow [Wed, 8 Oct 2008 11:26:27 +0000 (11:26 +0000)]
add comments and an ASSERT_LOCK_HELD()
Simon Marlow [Wed, 8 Oct 2008 11:26:09 +0000 (11:26 +0000)]
Fix #2663: we had a hard-wired capabilities[0]
For some unknown reason in schedulePostRunThread() we were always
passing capabilities[0] rather than the current Capability to
throwToSingleThreaded(). This caused all kinds of weird failures and
crashes in STM code when running on multiple processors.
Simon Marlow [Fri, 3 Oct 2008 15:07:45 +0000 (15:07 +0000)]
Fix #1955 for heap profiles generated by +RTS -hT
Simon Marlow [Tue, 7 Oct 2008 15:10:07 +0000 (15:10 +0000)]
add a section id for +RTS -hT
Simon Marlow [Tue, 7 Oct 2008 15:09:57 +0000 (15:09 +0000)]
update documentation for PostfixOperators
Simon Marlow [Tue, 7 Oct 2008 15:09:43 +0000 (15:09 +0000)]
fix markup
Roman Leshchinskiy [Wed, 8 Oct 2008 10:16:18 +0000 (10:16 +0000)]
Fix bug in DPH docs
Roman Leshchinskiy [Wed, 8 Oct 2008 06:47:54 +0000 (06:47 +0000)]
Add short DPH section to users guide
MERGE TO 6.10
Manuel M T Chakravarty [Wed, 8 Oct 2008 06:19:27 +0000 (06:19 +0000)]
Users Guide: added type family documentation
MERGE TO 6.10
Roman Leshchinskiy [Wed, 8 Oct 2008 03:28:59 +0000 (03:28 +0000)]
Track changes to package dph
MERGE TO 6.10
Ian Lynagh [Tue, 7 Oct 2008 15:23:18 +0000 (15:23 +0000)]
Build a profiled GHC API by default if p is in GhcLibWays
Ian Lynagh [Tue, 7 Oct 2008 14:48:55 +0000 (14:48 +0000)]
Check whether mk/validate.mk defines anything after validating
Ian Lynagh [Mon, 6 Oct 2008 10:19:59 +0000 (10:19 +0000)]
Remove #define _BSD_SOURCE from Stg.h
It's no longer needed, as base no longer #includes it
Thomas Schilling [Tue, 7 Oct 2008 13:57:05 +0000 (13:57 +0000)]
Make ghctags compile again.
Clemens Fruhwirth [Tue, 7 Oct 2008 10:04:17 +0000 (10:04 +0000)]
Revert AutoLinkPackages change for dynamic libraries. Cabal handles that now.
Clemens Fruhwirth [Tue, 7 Oct 2008 10:07:50 +0000 (10:07 +0000)]
Change suffix for dyn. linked executables from _real to .dyn
Thomas Schilling [Mon, 6 Oct 2008 22:29:40 +0000 (22:29 +0000)]
Add accessors to 'Target' fields and haddockify.
MERGE TO 6.10
Thomas Schilling [Mon, 6 Oct 2008 22:28:31 +0000 (22:28 +0000)]
Make 'gblock' and 'gunblock' part of 'ExceptionMonad'. This way the
default implementations of 'gbracket' and 'gfinally' just work.
MERGE TO 6.10
Roman Leshchinskiy [Tue, 7 Oct 2008 00:44:16 +0000 (00:44 +0000)]
Add Word8 support to vectoriser
MERGE TO 6.10
Ian Lynagh [Sun, 5 Oct 2008 22:27:15 +0000 (22:27 +0000)]
Fix generating OS X installers: Set COMMAND_MODE=unix2003
If we don't specify COMMAND_MODE=unix2003 then xcodebuild defaults
to setting it to legacy, which means that ar builds archives
without a table of contents. That makes the build fail later on.
Ian Lynagh [Sun, 5 Oct 2008 14:33:07 +0000 (14:33 +0000)]
We need to set datadir = $(libdir) in bindists
We already do in the normal Makefiles.
This is because GHC needs package.conf and unlit to be in the same place
(and things like ghc-pkg need to agree on where package.conf is, so we
just set it globally).
Ian Lynagh [Sat, 4 Oct 2008 17:53:51 +0000 (17:53 +0000)]
prep-bin-dist-mingw complains if it finds a bad version of windres
'Jose Pedro Magalhaes [Thu, 2 Oct 2008 08:28:08 +0000 (08:28 +0000)]
removed Data.Generics.Basics, added Data.Data
Ian Lynagh [Fri, 3 Oct 2008 19:57:00 +0000 (19:57 +0000)]
Fix a build problem with GHC 6.4.2
Clemens Fruhwirth [Fri, 3 Oct 2008 18:53:04 +0000 (18:53 +0000)]
No AutoLinkPackages for dynamic library linking
Clemens Fruhwirth [Wed, 1 Oct 2008 12:56:48 +0000 (12:56 +0000)]
use ghcError for error in command line
simonpj@microsoft.com [Fri, 3 Oct 2008 17:12:07 +0000 (17:12 +0000)]
Fix warnings
Ian Lynagh [Fri, 3 Oct 2008 16:12:47 +0000 (16:12 +0000)]
Always use extensible exceptions in ghc-pkg, rather than using ifdefs
Ian Lynagh [Fri, 3 Oct 2008 16:01:29 +0000 (16:01 +0000)]
Use a proper exception for IOEnvFailure, not just a UserError
Ian Lynagh [Fri, 3 Oct 2008 14:02:16 +0000 (14:02 +0000)]
Use an extensible-exceptions package when bootstrapping
Ifdefs for whether we had extensible exceptions or not were spreading
through GHC's source, and things would only have got worse for the next
2-3 years, so instead we now use an implementation of extensible
exceptions built on top of the old exception type.
simonpj@microsoft.com [Fri, 3 Oct 2008 14:04:23 +0000 (14:04 +0000)]
Expunge ThFake, cure Trac #2632
This patch fixes a dirty hack (the fake ThFake module), which in turn
was causing Trac #2632.
The new scheme is that the top-level binders in a TH [d| ... |] decl splice
get Internal names. That breaks a previous invariant that things like
TyCons always have External names, but these TyCons are never long-lived;
they live only long enough to typecheck the TH quotation; the result is
discarded. So it seems cool.
Nevertheless -- Template Haskell folk: please test your code. The testsuite
is OK but it's conceivable that I've broken something in TH. Let's see.
simonpj@microsoft.com [Fri, 3 Oct 2008 14:01:44 +0000 (14:01 +0000)]
Make a debug check more refined
simonpj@microsoft.com [Fri, 3 Oct 2008 13:53:34 +0000 (13:53 +0000)]
Add ASSERTs to all calls of nameModule
nameModule fails on an InternalName. These ASSERTS tell you
which call failed.
Thomas Schilling [Thu, 2 Oct 2008 23:04:12 +0000 (23:04 +0000)]
Let parseModule take a ModSummary like checkAndLoadModule did.
To get the ModSummary for a ModuleName getModSummary can be used.
It's not called find* or lookup* because it assumes that the module is
in the module graph and throws an exception if it cannot be found.
Overall, I'm not quite sure about the usefulness of this function
since the user has no control about which filetype to grab (hs or
hs-boot).
Ian Lynagh [Thu, 2 Oct 2008 20:15:19 +0000 (20:15 +0000)]
Remove some out-of-date entries from .darcs-boring
Manuel M T Chakravarty [Thu, 2 Oct 2008 13:45:39 +0000 (13:45 +0000)]
TFs: Allow repeated variables in left-hand sides of instances
MERGE TO 6.10
Manuel M T Chakravarty [Thu, 2 Oct 2008 07:46:42 +0000 (07:46 +0000)]
Clean up some comments
MERGE TO 6.10
simonpj@microsoft.com [Thu, 2 Oct 2008 13:30:02 +0000 (13:30 +0000)]
Make the new binder-swap stuff in OccurAnal work right for GlobalIds
See Note [Binder swap on GlobalId scrutinees]. I hadn't got this
right before, so repeated cases on imported Ids weren't getting optimised.
simonpj@microsoft.com [Thu, 2 Oct 2008 13:29:29 +0000 (13:29 +0000)]
Minor refactoring only
simonpj@microsoft.com [Thu, 2 Oct 2008 13:28:33 +0000 (13:28 +0000)]
Comments only
simonpj@microsoft.com [Thu, 2 Oct 2008 13:26:57 +0000 (13:26 +0000)]
Zap dead-ness info appropriately in SpecConstr
SpecConstr can make pattern binders come alive, so we must remember
to zap their dead-variable annotation. See extendCaseBndrs.
(This was triggering a Core Lint failure in DPH.)
simonpj@microsoft.com [Thu, 2 Oct 2008 13:24:26 +0000 (13:24 +0000)]
Suppress invalid Core Lint complaint about lack of constructors
Simon Marlow [Wed, 1 Oct 2008 16:44:27 +0000 (16:44 +0000)]
add some more GC roots (fixes conc048, and possibly some others)
Simon Marlow [Wed, 1 Oct 2008 16:32:22 +0000 (16:32 +0000)]
Document +RTS -hT
We forgot to document this in GHC 6.8
Simon Marlow [Wed, 1 Oct 2008 16:31:05 +0000 (16:31 +0000)]
fix new-qualified-operators link
Manuel M T Chakravarty [Wed, 1 Oct 2008 14:43:39 +0000 (14:43 +0000)]
Proper error message for unsupported pattern signatures
- Pattern signatures must be identical to the type expected for the pattern;
see Note [Pattern coercions]
- We now signal an appropriate error if an equality coercion would be needed
(instead of just generating Core that doesn't typecheck)
MERGE TO 6.10
Roman Leshchinskiy [Thu, 2 Oct 2008 01:20:55 +0000 (01:20 +0000)]
Prevent excessive inlining with DPH
This adds a new flag -finline-if-enough-args which disables inlining for
partially applied functions. It is automatically set by -Odph. This is a
temporary hack and should remain undocumented.
MERGE TO 6.10
Ian Lynagh [Wed, 1 Oct 2008 17:11:33 +0000 (17:11 +0000)]
On Windows, check that we have a good version of windres when configuring
Ian Lynagh [Tue, 30 Sep 2008 15:50:54 +0000 (15:50 +0000)]
Call $(PERL) rather than perl when making the manpage
Ian Lynagh [Tue, 30 Sep 2008 14:57:14 +0000 (14:57 +0000)]
don't install the installPackage program
Simon Marlow [Wed, 1 Oct 2008 13:55:49 +0000 (13:55 +0000)]
Fix #2637: conc032(threaded2) failure
There was a race condition whereby a thread doing throwTo could be
blocked on a thread that had finished, and the GC would detect this
as a deadlock rather than raising the pending exception. We can't
close the race, but we can make the right thing happen when the GC
runs later.
Simon Marlow [Tue, 30 Sep 2008 15:09:12 +0000 (15:09 +0000)]
Remove outdated link to OGI webpage
Manuel M T Chakravarty [Wed, 1 Oct 2008 13:13:03 +0000 (13:13 +0000)]
TFs: Fixed InstContextNorm (and simplification of IPs)
MERGE TO 6.10
Manuel M T Chakravarty [Wed, 1 Oct 2008 09:13:15 +0000 (09:13 +0000)]
TcSimplify.reduceImplication: clean up
- This cleans up some of the mess in reduceImplication and documents the
precondition on the form of wanted equalities properly.
- I also made the back off test a bit smarter by allowing to back off in the
presence of wanted equalities as long as none of them got solved in the
attempt. (That should save generating some superfluous bindings.)
MERGE TO 6.10
Manuel M T Chakravarty [Wed, 1 Oct 2008 05:32:43 +0000 (05:32 +0000)]
Make sure to zonk the kind of coercion variables
MERGE TO 6.10
Simon Marlow [Tue, 30 Sep 2008 14:18:42 +0000 (14:18 +0000)]
Remover PROT_EXEC flag from mmap()
Needed for #738 fix
Simon Marlow [Tue, 30 Sep 2008 14:18:12 +0000 (14:18 +0000)]
Fix #2410: carefully generate unique names for CAF CCs
Simon Marlow [Tue, 30 Sep 2008 11:56:11 +0000 (11:56 +0000)]
fix #2594: we were erroneously applying masks, as the reporter suggested
My guess is that this is left over from when we represented Int8 and
friends as zero-extended rather than sign-extended. It's amazing it hasn't
been noticed earlier.
Clemens Fruhwirth [Tue, 30 Sep 2008 08:54:49 +0000 (08:54 +0000)]
Unconditionalize definition of DYNAMIC_* so that libffi.so/.dll is removed even when BuildSharedLibs is reset to NO
Manuel M T Chakravarty [Tue, 30 Sep 2008 05:35:59 +0000 (05:35 +0000)]
Type families: need to instantiate flexible skolems before other flexibles
MERGE TO 6.10
Manuel M T Chakravarty [Mon, 29 Sep 2008 14:22:27 +0000 (14:22 +0000)]
Fix warnings
Manuel M T Chakravarty [Mon, 29 Sep 2008 14:10:40 +0000 (14:10 +0000)]
Type families: consider subst rules both way
- applySubstFam, applySubstVarVar & applySubstVarFam need to return their
second argument -to be put into the todo list- if the rule would be
applicable if the equalities would be supplied in the opposite order.
MERGE TO 6.10
pepe [Fri, 26 Sep 2008 21:14:29 +0000 (21:14 +0000)]
Clean up a bit and improve an error message
pepe [Fri, 26 Sep 2008 20:48:36 +0000 (20:48 +0000)]
Don't capture error calls in tryUser
A previous patch slightly changed the semantics of tryUser.
This patch restores the original behaviour
(as expected in :print)