Simon Marlow [Fri, 22 Jun 2007 07:54:44 +0000 (07:54 +0000)]
FIX BUILD: all builds need --template, not just Windows
Simon Marlow [Fri, 22 Jun 2007 07:54:17 +0000 (07:54 +0000)]
ignore all but the last --template option
simonpj@microsoft.com [Thu, 21 Jun 2007 17:01:31 +0000 (17:01 +0000)]
Unbreak the stage-2 compiler (record-type changes)
simonpj@microsoft.com [Thu, 21 Jun 2007 16:21:52 +0000 (16:21 +0000)]
-fglasgow-exts implies -X=GADTs
simonpj@microsoft.com [Thu, 21 Jun 2007 14:37:21 +0000 (14:37 +0000)]
FIX read040: patterns with type sig on LHS of do-binding
f () = do { x :: Bool <- return True; ... }
For some reason the production for 'pat' required 'infixexp' on the
LHS of a do-notation binding. This patch makes it an 'exp', which
thereby allows an expression with a type sig.
Happily, there are no new shift-reduce errors, so I don't think this
will break anything else.
simonpj@microsoft.com [Thu, 21 Jun 2007 14:11:44 +0000 (14:11 +0000)]
Another wibble to the head-exploded error message (suggested by David Roundy)
simonpj@microsoft.com [Thu, 21 Jun 2007 14:11:14 +0000 (14:11 +0000)]
Use the correct flag for controlling scoped type variables in an instance decl
simonpj@microsoft.com [Thu, 21 Jun 2007 13:16:50 +0000 (13:16 +0000)]
Improve 'my head exploded' error message
simonpj@microsoft.com [Thu, 21 Jun 2007 13:16:25 +0000 (13:16 +0000)]
FIX BUILD: add missing prime!
Simon Marlow [Thu, 21 Jun 2007 14:45:19 +0000 (14:45 +0000)]
FIX BUILD on Windows: horrible hack to work around make(?) bug
Simon Marlow [Thu, 21 Jun 2007 14:06:53 +0000 (14:06 +0000)]
yet more fixes: Cygwin broke this time
Lemmih [Thu, 21 Jun 2007 13:28:25 +0000 (13:28 +0000)]
fix bugs with hsc2hs-inplace
Lemmih [Thu, 21 Jun 2007 13:07:27 +0000 (13:07 +0000)]
mk/build.mk is optional
Lemmih [Thu, 21 Jun 2007 13:06:57 +0000 (13:06 +0000)]
further fixes to the inplace scripts
Simon Marlow [Thu, 21 Jun 2007 12:14:26 +0000 (12:14 +0000)]
use a binary for hsc2hs-inplace too
Simon Marlow [Thu, 21 Jun 2007 11:41:47 +0000 (11:41 +0000)]
Fix problems with new inplace stuff on Cygwin
Simon Marlow [Thu, 21 Jun 2007 09:51:54 +0000 (09:51 +0000)]
FIX BUILD: can't build the makefiles during make boot
Simon Marlow [Wed, 20 Jun 2007 13:28:09 +0000 (13:28 +0000)]
clean up lib/{GNUmakefile,Makefile.local}
Simon Marlow [Wed, 20 Jun 2007 13:20:01 +0000 (13:20 +0000)]
default_target should be "all", not "build"
Simon Marlow [Wed, 20 Jun 2007 13:19:49 +0000 (13:19 +0000)]
add comment
Simon Marlow [Wed, 20 Jun 2007 12:22:24 +0000 (12:22 +0000)]
Fix a problem with package.mk being included too early
Simon Marlow [Wed, 20 Jun 2007 12:20:09 +0000 (12:20 +0000)]
Use setup makefile + make by default to build libraries
The advantages of this are
(a) it's a step closer to getting -j working again (make -j works in
an individual library, but not in libraries/ yet).
(b) it's easier to hack on libraries: make dist/build/Foo.o
(c) it's a step closer to getting HC bootstrapping again
The build system creates <lib>/GNUmakefile as part of 'make boot'.
This was chosen so as not to interfere with existing Makefiles, but
it's a bit of a hack. (previously I used CabalMakefile, but that
means adding -f CabalMakefile each time you run make, and that's a
pain).
Simon Marlow [Thu, 21 Jun 2007 10:13:24 +0000 (10:13 +0000)]
Use a real binary instead of scripts for ghc-inplace
Fixes various problems with getting the scripts right on Windows.
Binaries are universally executable by /bin/sh, cmd.exe and rawSystem,
so this allows us to remove some platform-specific hacks.
Lemmih [Thu, 21 Jun 2007 09:15:52 +0000 (09:15 +0000)]
Add several new record features
1. Record disambiguation (-fdisambiguate-record-fields)
In record construction and pattern matching (although not
in record updates) it is clear which field name is intended
even if there are several in scope. This extension uses
the constructor to disambiguate. Thus
C { x=3 }
uses the 'x' field from constructor C (assuming there is one)
even if there are many x's in scope.
2. Record punning (-frecord-puns)
In a record construction or pattern match or update you can
omit the "=" part, thus
C { x, y }
This is just syntactic sugar for
C { x=x, y=y }
3. Dot-dot notation for records (-frecord-dot-dot)
In record construction or pattern match (but not update)
you can use ".." to mean "all the remaining fields". So
C { x=v, .. }
means to fill in the remaining fields to give
C { x=v, y=y }
(assuming C has fields x and y). This might reasonably
considered very dodgy stuff. For pattern-matching it brings
into scope a bunch of things that are not explictly mentioned;
and in record construction it just picks whatver 'y' is in
scope for the 'y' field. Still, Lennart Augustsson really
wants it, and it's a feature that is extremely easy to explain.
Implementation
~~~~~~~~~~~~~~
I thought of using the "parent" field in the GlobalRdrEnv, but
that's really used for import/export and just isn't right for this.
For example, for import/export a field is a subordinate of the *type
constructor* whereas here we need to know what fields belong to a
particular *data* constructor.
The main thing is that we need to map a data constructor to its
fields, and we need to do so in the renamer. For imported modules
it's easy: just look in the imported TypeEnv. For the module being
compiled, we make a new field tcg_field_env in the TcGblEnv.
The important functions are
RnEnv.lookupRecordBndr
RnEnv.lookupConstructorFields
There is still a significant infelicity in the way the renamer
works on patterns, which I'll tackle next.
I also did quite a bit of refactoring in the representation of
record fields (mainly in HsPat).***END OF DESCRIPTION***
Place the long patch description above the ***END OF DESCRIPTION*** marker.
The first line of this file will be the patch name.
This patch contains the following changes:
M ./compiler/deSugar/Check.lhs -3 +5
M ./compiler/deSugar/Coverage.lhs -6 +7
M ./compiler/deSugar/DsExpr.lhs -6 +13
M ./compiler/deSugar/DsMeta.hs -8 +8
M ./compiler/deSugar/DsUtils.lhs -1 +1
M ./compiler/deSugar/MatchCon.lhs -2 +2
M ./compiler/hsSyn/Convert.lhs -3 +3
M ./compiler/hsSyn/HsDecls.lhs -9 +25
M ./compiler/hsSyn/HsExpr.lhs -13 +3
M ./compiler/hsSyn/HsPat.lhs -25 +63
M ./compiler/hsSyn/HsUtils.lhs -3 +3
M ./compiler/main/DynFlags.hs +6
M ./compiler/parser/Parser.y.pp -13 +17
M ./compiler/parser/RdrHsSyn.lhs -16 +18
M ./compiler/rename/RnBinds.lhs -2 +2
M ./compiler/rename/RnEnv.lhs -22 +82
M ./compiler/rename/RnExpr.lhs -34 +12
M ./compiler/rename/RnHsSyn.lhs -3 +2
M ./compiler/rename/RnSource.lhs -50 +78
M ./compiler/rename/RnTypes.lhs -50 +84
M ./compiler/typecheck/TcExpr.lhs -18 +18
M ./compiler/typecheck/TcHsSyn.lhs -20 +21
M ./compiler/typecheck/TcPat.lhs -8 +6
M ./compiler/typecheck/TcRnMonad.lhs -6 +15
M ./compiler/typecheck/TcRnTypes.lhs -2 +11
M ./compiler/typecheck/TcTyClsDecls.lhs -3 +4
M ./docs/users_guide/flags.xml +7
M ./docs/users_guide/glasgow_exts.xml +42
Lemmih [Mon, 18 Jun 2007 12:46:05 +0000 (12:46 +0000)]
Remove the unused HsExpr constructor DictPat
Lemmih [Thu, 7 Jun 2007 21:38:37 +0000 (21:38 +0000)]
Fix a bug in MatchCon, and clarify what dataConInstOrigArgTys does
There was an outright bug in MatchCon.matchOneCon, in the construction
of arg_tys. Easily fixed. It never showed up becuase the arg_tys are
only used in WildPats, and they in turn seldom have their types looked
(except by hsPatType). So I can't make a test case for htis.
While I was investigating, I added a bit of clarifation and
invariant-checking to dataConInstOrigArgTys and dataConInstArgTys
Lemmih [Thu, 7 Jun 2007 21:35:23 +0000 (21:35 +0000)]
Do not perform a worker/wrapper split for a NOINLINE function
This came up in an email exchange with Duncan Coutts in May 2007.
If a function is marked NOINLINE there is really no point in
doing a worker/wrapper split, because the wrapper will never
be inlined.
Lemmih [Thu, 7 Jun 2007 18:55:34 +0000 (18:55 +0000)]
Fix Trac #1402: typo in specialiser
This patch fixes a plain bug in the specialiser (rhs_bndrs instead
of rhs_ids) which made GHC crash in obscure cases.
It exposed a case in which we might not do all possible specialisation;
see Note [Specialisation shape]. It's not an important case, but I've
added a warning in DEBUG mode.
Trac #1402. Test is spec003.hs
simonpj@microsoft.com [Thu, 21 Jun 2007 07:24:56 +0000 (07:24 +0000)]
Wibble: make -fno-implicit-prelude work
Ian Lynagh [Thu, 21 Jun 2007 00:31:03 +0000 (00:31 +0000)]
Remove an incorrect claim that [t| ... |] isn't implemented yet
Ian Lynagh [Wed, 20 Jun 2007 23:59:09 +0000 (23:59 +0000)]
Make building haddock docs opt-in rather than opt-out
Ian Lynagh [Wed, 20 Jun 2007 23:40:07 +0000 (23:40 +0000)]
Don't assume that the main repo is called "ghc" in darcs-all
Fixes working in branches where that isn't true.
simonpj@microsoft.com [Wed, 20 Jun 2007 16:58:36 +0000 (16:58 +0000)]
Trivial fix to clear Trac #1386
simonpj@microsoft.com [Wed, 20 Jun 2007 16:58:01 +0000 (16:58 +0000)]
Wibbles in flaggery, concerning backward compatibility with -f flags
simonpj@microsoft.com [Wed, 20 Jun 2007 16:33:59 +0000 (16:33 +0000)]
Implement -X=GADTs and -X=RelaxedPolyRec
Two new -X flags, one for GADTs and one for relaxed polymorphic recursion
This also fixes a rather confusing error message that the Darcs folk
tripped over.
simonpj@microsoft.com [Wed, 20 Jun 2007 16:26:56 +0000 (16:26 +0000)]
Use -X for language extensions
We've often talked about having a separate flag for language extensions,
and now we have one. You can say
-XImplicitParams
-X=ImplicitParams
-Ximplicit-params
as you like. These replace the "-f" flags with similar names (though
the -f prefix will serve as a synonym for -X for a while).
There's an optional "=", and the flag is normalised by removing hyphens
and lower-casing, so all the above variants mean the same thing.
The nomenclature is intended to match the LANGUAGE pramgas, which are
defined by Cabal. So you can also say
{-# LANGUAGE ImplicitParams #-}
But Cabal doesn't have as many language options as GHC does, so the -X
things are a superset of the LANGUAGE things.
The optional "=" applies to all flags that take an argument, so you can,
for example, say
-pgmL=/etc/foo
I hope that's ok. (It's an unforced change; just fitted in.)
I hope we'll add more -X flags, to replace the portmanteau -fglasgow-exts
which does everything!
I have updated the manual, but doubtless missed something.
simonpj@microsoft.com [Wed, 20 Jun 2007 16:13:41 +0000 (16:13 +0000)]
Unused import
andy@galois.com [Wed, 20 Jun 2007 15:02:44 +0000 (15:02 +0000)]
turning back on case liberation when using hpc
Simon Marlow [Wed, 20 Jun 2007 12:57:42 +0000 (12:57 +0000)]
remove debugging code accidentally left in
Ian Lynagh [Wed, 20 Jun 2007 12:07:23 +0000 (12:07 +0000)]
Use .NOTPARALLEL for the libraries Makefile
simonpj@microsoft.com [Wed, 20 Jun 2007 10:28:28 +0000 (10:28 +0000)]
Improve the handling of deriving, in error cases
I'd been too ambitious with error handling for 'deriving', and got it
entirely wrong. This fixes it. See extensive
Note [Exotic derived instance contexts]
in TcSimplify. (Most of the extra lines are comments!)
simonpj@microsoft.com [Wed, 20 Jun 2007 09:36:46 +0000 (09:36 +0000)]
More refactoring in TcSimplify
This re-jig tides up the top-level simplification, and combines in one
well-commented function, approximateImplications, the rather ad-hoc
way of simplifying implication constraints during type inference.
Error messages get a bit better too.
simonpj@microsoft.com [Wed, 20 Jun 2007 07:56:48 +0000 (07:56 +0000)]
Fix egregious sharing bug in LiberateCase
Andy Gill writes: consider the following code
f = g (case v of
V a b -> a : t f)
where g is expensive. Liberate case will turn this into
f = g (case v of
V a b -> a : t (letrec f = g (case v of
V a b -> a : f t)
in f)
)
Yikes! We evaluate g twice. This leads to a O(2^n) explosion
if g calls back to the same code recursively.
This may be the same as Trac #1366.
andy@galois.com [Wed, 20 Jun 2007 06:49:43 +0000 (06:49 +0000)]
Turning off case liberation when using the hpc option, for now
Consider the following code
f = g (case v of
V a b -> a : t f)
where g is expensive. Liberate case will turn this into
f = g (case v of
V a b -> a : t (letrec f = g (case v of
V a b -> a : f t)
in f)
)
Yikes! We evaluate g twice. This leads to a O(2^n) explosion
if g calls back to the same code recursively.
This happen sometimes in HPC, because every tick is a liberate-able case,
but is a general problem to case liberation (I think).
Ian Lynagh [Tue, 19 Jun 2007 20:05:46 +0000 (20:05 +0000)]
Add --core-only flag to push-all
Ian Lynagh [Tue, 19 Jun 2007 19:28:20 +0000 (19:28 +0000)]
Add a push-all script
simonpj@microsoft.com [Tue, 19 Jun 2007 16:53:54 +0000 (16:53 +0000)]
Improve misleading warning (Trac #1422)
simonpj@microsoft.com [Tue, 19 Jun 2007 16:26:13 +0000 (16:26 +0000)]
Fix a bug in the handling of implication constraints (Trac #1430)
Trac #1430 showed up quite a nasty bug in the handling of implication
constraints when we are *inferring* the type of a function.
See Note [Inference and implication constraints]:
We can't (or at least don't) abstract over implications. But we might
have an implication constraint (perhaps arising from a nested pattern
match) like
C a => D a
when we are now trying to quantify over 'a'. Our best approximation
is to make (D a) part of the inferred context, so we can use that to
discharge the implication. Hence getImplicWanteds.
My solution is not marvellous, but it's better than before. I transferred
function getDefaultableDicts from Inst to TcSimplify (since it's only
called there). Many of the remaining 50 new lines are comments. But
there is undoubtedly more code than before (sigh).
Test is tc228.
simonpj@microsoft.com [Tue, 19 Jun 2007 16:26:03 +0000 (16:26 +0000)]
Comments only
simonpj@microsoft.com [Tue, 19 Jun 2007 08:46:34 +0000 (08:46 +0000)]
Remove erroneous requirement to import Control.Monad.Fix when using mdo
See Trac #1426
andy@galois.com [Tue, 19 Jun 2007 05:56:54 +0000 (05:56 +0000)]
First cut at documentation for HPC option in GHC
Manuel M T Chakravarty [Tue, 19 Jun 2007 01:15:10 +0000 (01:15 +0000)]
Build package ndp if present
Simon Marlow [Mon, 18 Jun 2007 11:18:17 +0000 (11:18 +0000)]
typo
Ian Lynagh [Mon, 18 Jun 2007 10:28:50 +0000 (10:28 +0000)]
More debugger output order consistency
Pepe Iborra [Sun, 17 Jun 2007 19:34:35 +0000 (19:34 +0000)]
Several changes to the code dealing with newtypes in :print
I simplified the code, killed some unreachable blocks, and renamed it so that it corresponds more accurately with what is explained in the technical report
http://www.dsic.upv.es/docs/bib-dig/informes/etd-
04042007-111431/papernew2.pdf
Also, fixed a bug related to newtypes in the pretty printer
Pepe Iborra [Wed, 13 Jun 2007 09:21:02 +0000 (09:21 +0000)]
Remove now non-existant "Breakpoints" entry from package.conf.in
Ian Lynagh [Sun, 17 Jun 2007 21:52:05 +0000 (21:52 +0000)]
Sort names before printing them in the debugger so output order is consistent
Ian Lynagh [Sat, 16 Jun 2007 19:37:45 +0000 (19:37 +0000)]
Use %d rather than %zd on Windows
Ian Lynagh [Fri, 15 Jun 2007 18:45:27 +0000 (18:45 +0000)]
Add missing quotes in generated script
Ian Lynagh [Fri, 15 Jun 2007 18:23:37 +0000 (18:23 +0000)]
Fix size mismatch errors in mkDerivedConstants.c
Simon Marlow [Thu, 14 Jun 2007 09:57:27 +0000 (09:57 +0000)]
workaround for #1421 (Solaris linker being picky about .size)
Simon Marlow [Wed, 13 Jun 2007 14:45:05 +0000 (14:45 +0000)]
I didn't quite fix #1424 completely - hopefully this gets it right
Simon Marlow [Wed, 13 Jun 2007 14:24:31 +0000 (14:24 +0000)]
FIX #1424: x86_64 NCG generated wrong code for foreign call with >8 double args
I guess we have a missing test... I'll add one
Simon Marlow [Wed, 13 Jun 2007 11:15:52 +0000 (11:15 +0000)]
another fix for -hb: we appear to be freeing the hash table and arena twice
Simon Marlow [Wed, 13 Jun 2007 10:29:28 +0000 (10:29 +0000)]
FIX #1418 (partially)
When the con_desc field of an info table was made into a relative
reference, this had the side effect of making the profiling fields
(closure_desc and closure_type) also relative, but only when compiling
via C, and the heap profiler was still treating them as absolute,
leading to crashes when profiling with -hd or -hy.
This patch fixes up the story to be consistent: these fields really
should be relative (otherwise we couldn't make shared versions of the
profiling libraries), so I've made them relative and fixed up the RTS
to know about this.
Simon Marlow [Wed, 13 Jun 2007 09:52:01 +0000 (09:52 +0000)]
should be using GET_CON_DESC() to get the constructor name
Simon Marlow [Wed, 13 Jun 2007 09:51:44 +0000 (09:51 +0000)]
warning police
Simon Marlow [Wed, 13 Jun 2007 08:51:38 +0000 (08:51 +0000)]
Use $(if...) to get lazy tests instead of if..endif in a few places
This means it should be possible to set GhcUnregisterised=YES in
build.mk and the rest of the settings should follow automatically
(GhcWithNativeCodeGen, SplitObjs, GhcWithSMP).
Ian Lynagh [Tue, 12 Jun 2007 21:07:38 +0000 (21:07 +0000)]
Tweak banner printing
* -{short,long}-ghci-banner are now dynamic options, so you can put
":set -short-ghci-banner" in .ghci
* The -v2 banner information now always tells you what compiler booted GHC,
and what stage the compiler is. Thus we no longer assume that stage > 1
iff GHCI is defined.
cdsmith@twu.net [Fri, 1 Jun 2007 05:36:44 +0000 (05:36 +0000)]
FIX #1378 Add option for a shorter banner on GHCi startup
Add -short-ghci-banner and -long-ghci-banner. The default is long, which is
the current behavior. The short banner prints a one-line introduction with
only the version, web site, and ":? for help" message.
Ian Lynagh [Sat, 9 Jun 2007 15:14:38 +0000 (15:14 +0000)]
Windows bindist tweaking
Ian Lynagh [Sat, 9 Jun 2007 13:43:02 +0000 (13:43 +0000)]
Fix the ghc package in bindists
Ian Lynagh [Sat, 9 Jun 2007 11:22:35 +0000 (11:22 +0000)]
Fix up whitespace
Ian Lynagh [Sat, 9 Jun 2007 11:19:20 +0000 (11:19 +0000)]
Fix bindists on Windows
Ian Lynagh [Tue, 5 Jun 2007 13:39:56 +0000 (13:39 +0000)]
Install the RTS from a bindist correctly
Ian Lynagh [Mon, 4 Jun 2007 13:47:14 +0000 (13:47 +0000)]
Only add the extra Windows libraries if we are on Windows
Ian Lynagh [Mon, 4 Jun 2007 13:46:45 +0000 (13:46 +0000)]
Follow Cabal change; .setup-config is now dist/setup-config
Ian Lynagh [Sat, 2 Jun 2007 19:56:51 +0000 (19:56 +0000)]
Copy more of base's hacks into installPackage; *sigh*
Ian Lynagh [Sat, 2 Jun 2007 19:38:26 +0000 (19:38 +0000)]
Hack libsubdir so that it does the right thing on both Windows and non-Windows
Ian Lynagh [Sat, 2 Jun 2007 12:11:59 +0000 (12:11 +0000)]
Remove now-unused files
Ian Lynagh [Sat, 2 Jun 2007 12:10:17 +0000 (12:10 +0000)]
Fix ghcii.sh creation
Ian Lynagh [Sat, 2 Jun 2007 01:51:01 +0000 (01:51 +0000)]
Explicitly set datadir
Cabal defaults to the value we want on Linux, but uses
"C:\\Program Files\\Common Files" on Windows.
Ian Lynagh [Fri, 1 Jun 2007 22:43:01 +0000 (22:43 +0000)]
Fix a -- that somehow turned into a ----
Ian Lynagh [Fri, 1 Jun 2007 22:33:55 +0000 (22:33 +0000)]
Fix creation of hsc2hs-inplace.bat
Ian Lynagh [Fri, 1 Jun 2007 21:49:53 +0000 (21:49 +0000)]
ghc-pkg-inplace now has .bat extension on Windows
Ian Lynagh [Fri, 1 Jun 2007 21:05:55 +0000 (21:05 +0000)]
Add omitted files (mk/{install,recurse}.mk)
Ian Lynagh [Fri, 1 Jun 2007 19:06:59 +0000 (19:06 +0000)]
Fix bindist creation
Bindists should now work again, when doing "make install" at least.
"make in-place" is probably still broken.
Ian Lynagh [Thu, 31 May 2007 14:35:05 +0000 (14:35 +0000)]
Rework the build system a bit
Key changes:
* Always build as if BIN_DIST is 1. BIN_DIST is thus removed.
* Libraries are configured with prefix set to $$topdir rather than $(prefix)
Ian Lynagh [Wed, 30 May 2007 14:46:33 +0000 (14:46 +0000)]
Make package.conf files a bit more readable
andy@galois.com [Tue, 12 Jun 2007 07:46:55 +0000 (07:46 +0000)]
Adding new ffi calls into the Hpc rts subsystem
foreign import ccall unsafe hs_hpc_write :: CString -> IO ()
foreign import ccall unsafe hs_hpc_read :: CString -> IO ()
These write a Hpc description of the state of the world to a file,
or read a description into the current Hpc tickbox subsystem.
Simon Marlow [Tue, 12 Jun 2007 08:17:03 +0000 (08:17 +0000)]
don't add -threaded in stage2/3 if we didn't build the threaded RTS
Simon Marlow [Tue, 12 Jun 2007 07:46:57 +0000 (07:46 +0000)]
fix compile error in the !GHCI_TABLES_NEXT_TO_CODE case
Thorkil Naur [Fri, 18 May 2007 16:59:57 +0000 (16:59 +0000)]
Fix PPC Mac OS X memory access problem in SMP.h (#1362)
andy@galois.com [Sat, 9 Jun 2007 00:00:21 +0000 (00:00 +0000)]
Using blacklist of places not to cover, rather than reverse-engineer deriving.
andy@galois.com [Fri, 8 Jun 2007 22:40:16 +0000 (22:40 +0000)]
Not adding ticks to compiler generated derived code.
Manuel M T Chakravarty [Fri, 8 Jun 2007 06:10:46 +0000 (06:10 +0000)]
Fix worker/wrapper ids for newtype instances
- mkDataConIds forced an algebraic worker/wrapper on newtype instances (not
on vanilla newtypes). I am not sure where this came from (from code I
wrote or from the recent restructuring of MkIds), but its not the right
thing to do. In particular, it requires newtype instances to have con_info
labels in the generated code, which makes no sense.
- Worker construction for newtypes is smart enough that it will construct the
right signature for newtype instances (ie, one that mentions the family
type constructor, not the instance tycon).
Manuel M T Chakravarty [Fri, 8 Jun 2007 04:44:00 +0000 (04:44 +0000)]
Fix deriving of indexed data types
- Fix typo introduced when moving deriving-specific code from TcEnv to TcDeriv
- Thanks to Roman for the bug report
Ian Lynagh [Thu, 7 Jun 2007 13:55:46 +0000 (13:55 +0000)]
Add instructions for adding a library to a build tree
simonpj@microsoft.com [Wed, 6 Jun 2007 13:34:37 +0000 (13:34 +0000)]
Don't suggest -fno-monomorphism-restriction if it's already set
This patch implements the suggestion in Trac #1398. It's obviously
stupid to suggest -fno-monomorphism-restriction if the user is already
using it.
(Maybe another suggestion would be good, but this one clearly bogus.)
simonpj@microsoft.com [Wed, 6 Jun 2007 13:25:21 +0000 (13:25 +0000)]
Remove unnecessary free-variables from renamer
The renamer used to be responsible for making sure that all interfaces
with instance decls (other than orphans) were loaded. But TH makes that
impossible, so the typechecker does it, via checkWiredInTyCon.
This patch simply removes redundant additions to the free-variable set
in the renamer, which were there, I believe, solely to ensure that the
instances came in. Removing them should change nothing, but it's a
useful clean up.