ghc.git
11 years agolibffi now doesn't have an artificial make boot/all split
Ian Lynagh [Sat, 5 Jul 2008 15:50:25 +0000 (15:50 +0000)] 
libffi now doesn't have an artificial make boot/all split

11 years agoNeed to make all in gmp, not boot
Ian Lynagh [Sat, 5 Jul 2008 15:32:45 +0000 (15:32 +0000)] 
Need to make all in gmp, not boot

11 years agogmp didn't really fit into the make boot/all cycle, so don't try to force it
Ian Lynagh [Sat, 5 Jul 2008 14:03:54 +0000 (14:03 +0000)] 
gmp didn't really fit into the make boot/all cycle, so don't try to force it
Now we just run make in it at the start of the stage1 build

11 years agoBuild hsc2hs with Cabal
Ian Lynagh [Sat, 5 Jul 2008 13:42:08 +0000 (13:42 +0000)] 
Build hsc2hs with Cabal
This is very rough around teh edges at the moment.

11 years agoAdd a flag to disable linking with the haskell98 package
Ian Lynagh [Sat, 5 Jul 2008 13:41:15 +0000 (13:41 +0000)] 
Add a flag to disable linking with the haskell98 package

11 years agoUse the last compiler if more than one is specified
Ian Lynagh [Sat, 5 Jul 2008 12:14:26 +0000 (12:14 +0000)] 
Use the last compiler if more than one is specified

11 years agoImprove error messages from pwd
Ian Lynagh [Fri, 4 Jul 2008 23:33:43 +0000 (23:33 +0000)] 
Improve error messages from pwd

11 years agoIn utils/hsc2hs, add LICENSE and hsc2hs.cabal from the standalone repo
Ian Lynagh [Fri, 4 Jul 2008 22:22:06 +0000 (22:22 +0000)] 
In utils/hsc2hs, add LICENSE and hsc2hs.cabal from the standalone repo

11 years agoRemove fgl from the libraries Makefile
Ian Lynagh [Fri, 4 Jul 2008 22:10:26 +0000 (22:10 +0000)] 
Remove fgl from the libraries Makefile
It's no longer an extralib

11 years agoTell the bootstrapping Cabal where ghc-pkg is
Ian Lynagh [Fri, 4 Jul 2008 15:27:13 +0000 (15:27 +0000)] 
Tell the bootstrapping Cabal where ghc-pkg is

11 years agoFIX #2398: file locking wasn't thread-safe
Simon Marlow [Fri, 4 Jul 2008 14:46:26 +0000 (14:46 +0000)] 
FIX #2398: file locking wasn't thread-safe

11 years agoRemove out of date comments and point to the commentary
Simon Marlow [Fri, 20 Jun 2008 13:52:58 +0000 (13:52 +0000)] 
Remove out of date comments and point to the commentary
The wiki commentary is now the official description of recompilation
checking.

http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/RecompilationAvoidance

11 years agoRemove Cabal modules from compat
Ian Lynagh [Thu, 3 Jul 2008 22:46:33 +0000 (22:46 +0000)] 
Remove Cabal modules from compat
We now get them from the bootstrapping package.conf instead

11 years agoFix trac #2307: conflicting functional dependencies
Ian Lynagh [Thu, 3 Jul 2008 19:25:40 +0000 (19:25 +0000)] 
Fix trac #2307: conflicting functional dependencies
We were accepting some instances that should have been rejected as
their fundep constraints were violated. e.g. we accepted
    class C a b c | b -> c
    instance C Bool Int Float
    instance C Char Int Double

11 years agoIf we know where alex, haddock and happy are then tell Cabal; fixes trac #2373
Ian Lynagh [Thu, 3 Jul 2008 19:10:31 +0000 (19:10 +0000)] 
If we know where alex, haddock and happy are then tell Cabal; fixes trac #2373

11 years agoDon't clean bootstrapping bits when cleaning libraries
Ian Lynagh [Thu, 3 Jul 2008 15:46:47 +0000 (15:46 +0000)] 
Don't clean bootstrapping bits when cleaning libraries

11 years agoMore libraries/Makefile fixes
Ian Lynagh [Thu, 3 Jul 2008 14:10:16 +0000 (14:10 +0000)] 
More libraries/Makefile fixes

11 years agoShove the GHC path through cygpath -m
Ian Lynagh [Thu, 3 Jul 2008 13:26:14 +0000 (13:26 +0000)] 
Shove the GHC path through cygpath -m

11 years agoTweak the configure script Windows-specific bits
Ian Lynagh [Thu, 3 Jul 2008 13:24:37 +0000 (13:24 +0000)] 
Tweak the configure script Windows-specific bits

11 years agoUse cygpath -m, rather than fudging it ourselves with sed
Ian Lynagh [Thu, 3 Jul 2008 13:17:25 +0000 (13:17 +0000)] 
Use cygpath -m, rather than fudging it ourselves with sed

11 years agoFix build on Windows
Ian Lynagh [Thu, 3 Jul 2008 12:45:53 +0000 (12:45 +0000)] 
Fix build on Windows

11 years agoInclude ghc.spec in tarballs; patch from, and fixes, trac #2390
Ian Lynagh [Thu, 3 Jul 2008 16:14:57 +0000 (16:14 +0000)] 
Include ghc.spec in tarballs; patch from, and fixes, trac #2390

11 years agoAdd a program for describing unexpected tests in testlog
Ian Lynagh [Thu, 3 Jul 2008 13:40:03 +0000 (13:40 +0000)] 
Add a program for describing unexpected tests in testlog
This goes through the testlog and spits out any sections that contain
"unexpected".

11 years agoTeach cabal-bin how to build Setup programs
Ian Lynagh [Thu, 3 Jul 2008 00:13:00 +0000 (00:13 +0000)] 
Teach cabal-bin how to build Setup programs

We now build a copy of Cabal and put it in a bootstrapping package.conf.

We also make boot in libraries much earlier in the build process, so we
can use cabal-bin for more stuff in the future.

11 years agoWibble cabal-bin's error message
Ian Lynagh [Wed, 2 Jul 2008 15:59:37 +0000 (15:59 +0000)] 
Wibble cabal-bin's error message
We don't need to put the program name in it, as that happens automatically

11 years agoAdd type signatures
Roman Leshchinskiy [Thu, 3 Jul 2008 02:36:35 +0000 (02:36 +0000)] 
Add type signatures

11 years agoCommand-line options for selecting DPH backend
Roman Leshchinskiy [Wed, 2 Jul 2008 02:22:02 +0000 (02:22 +0000)] 
Command-line options for selecting DPH backend

It's -fdph-seq and -fdph-par at the moment, I'll think of a nicer setup later.

11 years agoAdd missing dph package to Makefile
Roman Leshchinskiy [Wed, 2 Jul 2008 02:21:42 +0000 (02:21 +0000)] 
Add missing dph package to Makefile

11 years agoSlight refactoring
Roman Leshchinskiy [Tue, 1 Jul 2008 02:52:27 +0000 (02:52 +0000)] 
Slight refactoring

11 years agoRename *NDP* -> *DPH*
Roman Leshchinskiy [Tue, 1 Jul 2008 02:45:59 +0000 (02:45 +0000)] 
Rename *NDP* -> *DPH*

11 years agoParametrise vectoriser with DPH package
Roman Leshchinskiy [Tue, 1 Jul 2008 02:45:15 +0000 (02:45 +0000)] 
Parametrise vectoriser with DPH package

11 years agoDon't use DPH backend directly in vectoriser
Roman Leshchinskiy [Tue, 1 Jul 2008 02:14:36 +0000 (02:14 +0000)] 
Don't use DPH backend directly in vectoriser

11 years agoMake dph-seq and dph-par wired-in packages
Roman Leshchinskiy [Tue, 1 Jul 2008 02:02:14 +0000 (02:02 +0000)] 
Make dph-seq and dph-par wired-in packages

11 years agoFIX #2313 do not include BFD symbols in RTS when the BFD library is not available...
Karel Gardas [Wed, 28 May 2008 09:31:39 +0000 (09:31 +0000)] 
FIX #2313 do not include BFD symbols in RTS when the BFD library is not available for linking

11 years agoAdd --slow (and --fast) options to validate
Ian Lynagh [Tue, 1 Jul 2008 17:59:27 +0000 (17:59 +0000)] 
Add --slow (and --fast) options to validate
slow mode is 14% slower than normal. It uses -DDEBUG for the stage 2
compiler, and -XGenerics for the stage 2 compiler and the libraries.
I believe that most of the slowdown is actually caused by -XGenerics
rather than -DDEBUG.

11 years agoFix Trac #2307: need to nub bad fundep reports
simonpj@microsoft.com [Tue, 1 Jul 2008 16:58:30 +0000 (16:58 +0000)] 
Fix Trac #2307: need to nub bad fundep reports

11 years agoEasy fix for Trac #2409
simonpj@microsoft.com [Tue, 1 Jul 2008 16:37:22 +0000 (16:37 +0000)] 
Easy fix for Trac #2409

Yurgh. See Note [Desugaring seq (3)]

11 years agoMake a "validate --hpc"; shows how much of the compiler the testsuite tests
Ian Lynagh [Tue, 1 Jul 2008 12:45:15 +0000 (12:45 +0000)] 
Make a "validate --hpc"; shows how much of the compiler the testsuite tests
Currently it causes a load of ghci-debugger tests to fail and takes
63% longer.

11 years agoAllow the exact HPC tix filename to be given in the HPCTIXFILE env var
Ian Lynagh [Tue, 1 Jul 2008 12:43:20 +0000 (12:43 +0000)] 
Allow the exact HPC tix filename to be given in the HPCTIXFILE env var

11 years agoarray is now warning-free
Ian Lynagh [Mon, 30 Jun 2008 20:41:26 +0000 (20:41 +0000)] 
array is now warning-free

11 years agoSeveral fixes to 'deriving' including Trac #2378
simonpj@microsoft.com [Tue, 1 Jul 2008 12:09:08 +0000 (12:09 +0000)] 
Several fixes to 'deriving' including Trac #2378

This patch collects several related things together.

* Refactor TcDeriv so that the InstInfo and the method bindings are renamed
  together.  This was messy before, and is cleaner now.  Fixes a bug caused
  by interaction between the "auxiliary bindings" (which were given
  Original names before), and stand-alone deriving (which meant that those
  Original names came from a different module). Now the names are purely
  local an ordinary.

  To do this, InstInfo is parameterised like much else HsSyn stuff.

* Improve the location info in a dfun, which in turn improves location
  info for error messages, e.g. overlapping instances

* Make sure that newtype-deriving isn't used for Typeable1 and friends.
  (Typeable was rightly taken care of, but not Typeable1,2, etc.)

* Check for data types in deriving Data, so that you can't do, say,
  deriving instance Data (IO a)

* Decorate the derived binding with location info from the *instance*
  rather than from the *tycon*.  Again, this really only matters with
  standalone deriving, but it makes a huge difference there.

I think that's it.  Quite a few error messages change slightly.

If we release 6.8.4, this should go in if possible.

11 years agoFollow Cabal changes
Ian Lynagh [Sun, 29 Jun 2008 21:16:33 +0000 (21:16 +0000)] 
Follow Cabal changes

11 years agoRename cabal to cabal-bin
Ian Lynagh [Sun, 29 Jun 2008 11:00:03 +0000 (11:00 +0000)] 
Rename cabal to cabal-bin
Avoids conflicts with the Cabal library on case-insensitive filesystems

11 years agomkdirhier.sh now accepts -q, which makes it be quiet
Ian Lynagh [Fri, 27 Jun 2008 23:35:28 +0000 (23:35 +0000)] 
mkdirhier.sh now accepts -q, which makes it be quiet

11 years agoUpdate .darcs-boring
Ian Lynagh [Fri, 27 Jun 2008 18:14:10 +0000 (18:14 +0000)] 
Update .darcs-boring

11 years agoUpdate darcs-boring
Ian Lynagh [Fri, 27 Jun 2008 15:41:42 +0000 (15:41 +0000)] 
Update darcs-boring

11 years agoUpdate .darcs-boring
Ian Lynagh [Fri, 27 Jun 2008 14:56:33 +0000 (14:56 +0000)] 
Update .darcs-boring

11 years agoFollow Cabal changes
Ian Lynagh [Thu, 26 Jun 2008 20:27:49 +0000 (20:27 +0000)] 
Follow Cabal changes

11 years agoAbsolutify a path
Ian Lynagh [Thu, 26 Jun 2008 18:11:34 +0000 (18:11 +0000)] 
Absolutify a path
When building ghc-prim/Setup we weren't putting the hi files in the
right place.

11 years agoRemove fgl from extralibs
Ian Lynagh [Thu, 26 Jun 2008 11:27:39 +0000 (11:27 +0000)] 
Remove fgl from extralibs

11 years agoUse a program similar to cabal-install to build the libraries
Ian Lynagh [Thu, 26 Jun 2008 11:25:11 +0000 (11:25 +0000)] 
Use a program similar to cabal-install to build the libraries
This means that we don't have to make a Setup program for each library
individually, and also simplifies the build system a bit.

11 years agoFix Trac #2394: test for non-algebraic types in standalone deriving
simonpj@microsoft.com [Wed, 25 Jun 2008 16:02:04 +0000 (16:02 +0000)] 
Fix Trac #2394: test for non-algebraic types in standalone deriving

11 years ago() is now in ghc-prim:GHC.Unit
Ian Lynagh [Tue, 24 Jun 2008 14:48:49 +0000 (14:48 +0000)] 
() is now in ghc-prim:GHC.Unit

11 years agoGenerate a warning-free GHC.PrimopWrappers. ghc-prim is now -Wall clean.
Ian Lynagh [Tue, 24 Jun 2008 12:25:29 +0000 (12:25 +0000)] 
Generate a warning-free GHC.PrimopWrappers. ghc-prim is now -Wall clean.

11 years agoFix some inconsistencies in the code and docs of primitives
Ian Lynagh [Mon, 23 Jun 2008 22:34:54 +0000 (22:34 +0000)] 
Fix some inconsistencies in the code and docs of primitives
We were inconsistent about whether to use the name "MutArr#" or
"MutableArray#". Likewise ByteArr#/ByteArray# and
MutByteArr#/MutableByteArray#.

11 years agoFix the build with GHC 6.4
Ian Lynagh [Mon, 23 Jun 2008 14:44:26 +0000 (14:44 +0000)] 
Fix the build with GHC 6.4

11 years agoDon't rebuild things with the stage2 compiler
Ian Lynagh [Sun, 22 Jun 2008 13:46:13 +0000 (13:46 +0000)] 
Don't rebuild things with the stage2 compiler
It leads to annoying rebuilding when working in a built tree.
We'll handle this differently for 6.10.

11 years agoeditline is now warning-free
Ian Lynagh [Fri, 20 Jun 2008 21:21:10 +0000 (21:21 +0000)] 
editline is now warning-free

11 years agoRemove special handling for character types of characters >= 128, <= 255
Ian Lynagh [Sat, 21 Jun 2008 17:11:00 +0000 (17:11 +0000)] 
Remove special handling for character types of characters >= 128, <= 255
Many of the character types were wrong. Now the asc* names really do mean
ASCII, rather than latin-1.

11 years agoRemove code that isn't used now that we assume that GHC >= 6.4
Ian Lynagh [Fri, 20 Jun 2008 19:30:03 +0000 (19:30 +0000)] 
Remove code that isn't used now that we assume that GHC >= 6.4

11 years agoNow that we require GHC >= 6.4.2, System.IO.Error is always available
Ian Lynagh [Fri, 20 Jun 2008 19:10:59 +0000 (19:10 +0000)] 
Now that we require GHC >= 6.4.2, System.IO.Error is always available

11 years agohpc is -Wall clean
Ian Lynagh [Fri, 20 Jun 2008 14:20:58 +0000 (14:20 +0000)] 
hpc is -Wall clean

11 years agofilepath is now warning-free
Ian Lynagh [Fri, 20 Jun 2008 13:56:52 +0000 (13:56 +0000)] 
filepath is now warning-free

11 years agopretty is now -Wall clean
Ian Lynagh [Fri, 20 Jun 2008 13:53:35 +0000 (13:53 +0000)] 
pretty is now -Wall clean

11 years agoprocess is now -Wall clean
Ian Lynagh [Fri, 20 Jun 2008 01:18:32 +0000 (01:18 +0000)] 
process is now -Wall clean

11 years agodirectory is now -Wall clean
Ian Lynagh [Fri, 20 Jun 2008 01:13:35 +0000 (01:13 +0000)] 
directory is now -Wall clean

11 years agointeger-gmp is warning-free
Ian Lynagh [Thu, 19 Jun 2008 23:59:25 +0000 (23:59 +0000)] 
integer-gmp is warning-free

11 years agopackedstring is now -Wall clean
Ian Lynagh [Thu, 19 Jun 2008 23:56:12 +0000 (23:56 +0000)] 
packedstring is now -Wall clean

11 years agoold-time is now warning-free
Ian Lynagh [Thu, 19 Jun 2008 23:31:27 +0000 (23:31 +0000)] 
old-time is now warning-free

11 years agoold-locale is now warning-free
Ian Lynagh [Thu, 19 Jun 2008 23:21:52 +0000 (23:21 +0000)] 
old-locale is now warning-free

11 years agorandom is now -Wall clean
Ian Lynagh [Thu, 19 Jun 2008 14:02:11 +0000 (14:02 +0000)] 
random is now -Wall clean

11 years agodocument some RTS options I added a while ago: -qm, -qw
Simon Marlow [Thu, 19 Jun 2008 12:12:27 +0000 (12:12 +0000)] 
document some RTS options I added a while ago: -qm, -qw

11 years agoDocument the change to +RTS -S/-s/-t
Simon Marlow [Thu, 19 Jun 2008 12:12:01 +0000 (12:12 +0000)] 
Document the change to +RTS -S/-s/-t

11 years agodocument paralel GC option: +RTS -g
Simon Marlow [Thu, 19 Jun 2008 12:11:11 +0000 (12:11 +0000)] 
document paralel GC option: +RTS -g

11 years ago+RTS -S -RTS now sends output to stderr (also -s)
Simon Marlow [Thu, 19 Jun 2008 11:33:29 +0000 (11:33 +0000)] 
+RTS -S -RTS now sends output to stderr (also -s)
Previously +RTS -Sstderr -RTS was required to send output to stderr,
but this is the most common usage and I got tired of typing "stderr".
The old default behaviour of sending output to <prog>.stat is now gone
(I don't think we use it anywhere).

Temporarily we allowed +RTS -S- -RTS to mean stderr; there were
objections to this, so it is now also gone.

11 years agobytestring is -Wall clean
Ian Lynagh [Thu, 19 Jun 2008 01:07:02 +0000 (01:07 +0000)] 
bytestring is -Wall clean

11 years agoCabal is -Wall clean
Ian Lynagh [Thu, 19 Jun 2008 01:04:36 +0000 (01:04 +0000)] 
Cabal is -Wall clean

11 years agoThe haskell98 library is -Wall clean
Ian Lynagh [Thu, 19 Jun 2008 01:01:24 +0000 (01:01 +0000)] 
The haskell98 library is -Wall clean

11 years agotemplate-haskell is now -Wall clean
Ian Lynagh [Thu, 19 Jun 2008 00:58:11 +0000 (00:58 +0000)] 
template-haskell is now -Wall clean

11 years agocontainers is now -Wall clean
Ian Lynagh [Wed, 18 Jun 2008 23:36:51 +0000 (23:36 +0000)] 
containers is now -Wall clean

11 years agofix a tiny bug spotted by gcc 4.3
Simon Marlow [Thu, 19 Jun 2008 10:09:04 +0000 (10:09 +0000)] 
fix a tiny bug spotted by gcc 4.3

11 years agoFix up inlines for gcc 4.3
Simon Marlow [Thu, 19 Jun 2008 10:08:49 +0000 (10:08 +0000)] 
Fix up inlines for gcc 4.3
gcc 4.3 emits warnings for static inline functions that its heuristics
decided not to inline.  The workaround is to either mark appropriate
functions as "hot" (a new attribute in gcc 4.3), or sometimes to use
"extern inline" instead.

With this fix I can validate with gcc 4.3 on Fedora 9.

11 years agofix warnings with gcc 4.3
Simon Marlow [Wed, 18 Jun 2008 14:43:07 +0000 (14:43 +0000)] 
fix warnings with gcc 4.3

11 years agoit's time to retire ghcprof & friends
Simon Marlow [Wed, 18 Jun 2008 14:02:28 +0000 (14:02 +0000)] 
it's time to retire ghcprof & friends

11 years agoMake the wired-in packages code handle ndp mapping to ndp-seq or ndp-par
Ian Lynagh [Wed, 18 Jun 2008 16:22:33 +0000 (16:22 +0000)] 
Make the wired-in packages code handle ndp mapping to ndp-seq or ndp-par
It's getting a bit crufty - could probably do with a rewrite.

11 years agoPut the infrastructure in place for getting the libraries -Wall clean
Ian Lynagh [Tue, 17 Jun 2008 02:01:45 +0000 (02:01 +0000)] 
Put the infrastructure in place for getting the libraries -Wall clean
libraries/Makefile.local now lists those for which we need to pass -w
(currently this is every library).

11 years agoutils/Digraph doesn't need -fglasgow-exts passed specially
Ian Lynagh [Mon, 16 Jun 2008 22:59:49 +0000 (22:59 +0000)] 
utils/Digraph doesn't need -fglasgow-exts passed specially

11 years agoFix Trac #2321: bug in SAT
simonpj@microsoft.com [Mon, 16 Jun 2008 20:17:00 +0000 (20:17 +0000)] 
Fix Trac #2321: bug in SAT
  This is a fairly substantial rewrite of the Static Argument Transformatoin,
  done by Max Bolingbroke and reviewed and modified by Simon PJ.

  * Fix a subtle scoping problem; see Note [Binder type capture]
  * Redo the analysis to use environments
  * Run gentle simlification just before the transformation

11 years agodefine NeedVarargsPrototypes to avoid segfault on x86_64
Simon Marlow [Wed, 18 Jun 2008 13:21:16 +0000 (13:21 +0000)] 
define NeedVarargsPrototypes to avoid segfault on x86_64

11 years agoFix an example where we weren't doing case-of-case when we should
Simon Marlow [Tue, 17 Jun 2008 12:35:10 +0000 (12:35 +0000)] 
Fix an example where we weren't doing case-of-case when we should
That's 1 line of new code and 38 lines of new comments

11 years agofix gcc warnings for printf formats on 32-bit
Simon Marlow [Wed, 18 Jun 2008 09:40:18 +0000 (09:40 +0000)] 
fix gcc warnings for printf formats on 32-bit

11 years agoFIX BUILD on Windows
Simon Marlow [Wed, 18 Jun 2008 09:47:00 +0000 (09:47 +0000)] 
FIX BUILD on Windows

11 years agoSort the mi_deps into a canonical ordering before fingerprinting.
Simon Marlow [Tue, 17 Jun 2008 15:21:17 +0000 (15:21 +0000)] 
Sort the mi_deps into a canonical ordering before fingerprinting.
This may help do a little less recompilation with make (GHC's --make
is unaffected).

11 years agoFix another "urk! lookup local fingerprint" in nofib/real/bspt/GeomNum.lhs
Simon Marlow [Tue, 17 Jun 2008 15:15:30 +0000 (15:15 +0000)] 
Fix another "urk! lookup local fingerprint" in nofib/real/bspt/GeomNum.lhs

11 years agosmall interpreter fix
Simon Marlow [Tue, 17 Jun 2008 13:46:51 +0000 (13:46 +0000)] 
small interpreter fix

11 years agofix some printf formats for 64 bits
Simon Marlow [Tue, 17 Jun 2008 10:11:02 +0000 (10:11 +0000)] 
fix some printf formats for 64 bits

11 years ago64-bit fixes
Simon Marlow [Tue, 17 Jun 2008 10:10:45 +0000 (10:10 +0000)] 
64-bit fixes

11 years agodon't try to parallelise marking GC (yet)
Simon Marlow [Mon, 16 Jun 2008 07:31:11 +0000 (07:31 +0000)] 
don't try to parallelise marking GC (yet)

11 years agoanother stableptr003 fix
Simon Marlow [Mon, 9 Jun 2008 19:17:22 +0000 (19:17 +0000)] 
another stableptr003 fix

11 years agoExperimental "mark-region" strategy for the old generation
Simon Marlow [Mon, 9 Jun 2008 17:49:43 +0000 (17:49 +0000)] 
Experimental "mark-region" strategy for the old generation
Sometimes better than the default copying, enabled by +RTS -w