Duncan Coutts [Wed, 29 Apr 2009 14:29:40 +0000 (14:29 +0000)]
Don't actually build ghc itself with -dynamic
For now, with --enable-shared we'll build the libs shared,
but ghc itself will still be statically linked.
In any case, we would want to be able to build it both
ways so it's easy to test both.
Duncan Coutts [Tue, 28 Apr 2009 12:35:29 +0000 (12:35 +0000)]
Add a new $way_libsuf variable for library suffix+extension
This allows the library file type to depend on the way. This is needed
to use .so/.dll libs for the "dyn" way rather than always using .a libs.
For example: thr_debug_dyn_libsuf="_thr_debug-ghc6.11.
20090426.so"
Duncan Coutts [Tue, 28 Apr 2009 12:34:52 +0000 (12:34 +0000)]
extend the rules/c-objs macro to take the way as a parameter
Previously we only built library package "cbits" the vanilla way, afterall
C code does not need to be built differently for profiling builds. However
for dynamic libs the C code needs to be built with -fPIC, so we do need
to be able to build package .c (and .s?) files multiple ways.
Duncan Coutts [Tue, 28 Apr 2009 10:49:24 +0000 (10:49 +0000)]
Do not link ghc stage1 using -threaded, only for stage2 or 3
We link stage1 using the bootstrapping compiler and there's
no guarantee that it has working support for threaded
Ian Lynagh [Wed, 29 Apr 2009 18:37:28 +0000 (18:37 +0000)]
Tweak ghc-cabal
Ian Lynagh [Wed, 29 Apr 2009 18:19:09 +0000 (18:19 +0000)]
Add special support for haskeline
The library uses stuff in Setup.hs to determine whether or not it needs
to use -liconv. This patch replicates that logic in ghc-cabal. This
isn't pretty, and we should find a better way to do it, but it works
for now.
Ian Lynagh [Wed, 29 Apr 2009 15:12:15 +0000 (15:12 +0000)]
In mkdependC.prl, create temp files properly
Avoids a race condition, where one run deletes/overwrites the temp file
of another.
Ian Lynagh [Wed, 29 Apr 2009 00:58:38 +0000 (00:58 +0000)]
Use haskeline, rather than editline, for line editing in ghci
Ian Lynagh [Tue, 28 Apr 2009 23:28:48 +0000 (23:28 +0000)]
Fix error handling
After the fix to #2500, we could get "Failing due to -Werror." but no
warnings printed. Now we don't fail in that case.
Simon Marlow [Tue, 28 Apr 2009 12:55:51 +0000 (12:55 +0000)]
move nofib settings here from ghc's config.mk
Simon Marlow [Tue, 28 Apr 2009 12:55:28 +0000 (12:55 +0000)]
add missing eventlog subdir
Simon Marlow [Tue, 28 Apr 2009 12:45:05 +0000 (12:45 +0000)]
move runstdtest into nofib
Simon Marlow [Tue, 28 Apr 2009 08:53:45 +0000 (08:53 +0000)]
improve the finalizer callback error message
Simon Marlow [Tue, 28 Apr 2009 08:52:33 +0000 (08:52 +0000)]
Add a header to all build system files:
# -----------------------------------------------------------------------------
#
# (c) 2009 The University of Glasgow
#
# This file is part of the GHC build system.
#
# To understand how the build system works and how to modify it, see
# http://hackage.haskell.org/trac/ghc/wiki/Building/Architecture
# http://hackage.haskell.org/trac/ghc/wiki/Building/Modifying
#
# -----------------------------------------------------------------------------
Simon Marlow [Tue, 28 Apr 2009 08:49:19 +0000 (08:49 +0000)]
more cleaning
Simon Marlow [Mon, 27 Apr 2009 15:25:29 +0000 (15:25 +0000)]
todo updates
Simon Marlow [Mon, 27 Apr 2009 14:47:43 +0000 (14:47 +0000)]
remove commentary that is now in the wiki
Simon Marlow [Mon, 27 Apr 2009 14:44:06 +0000 (14:44 +0000)]
remove ticket todos
Simon Marlow [Mon, 27 Apr 2009 13:43:22 +0000 (13:43 +0000)]
require gcc 3.0+ (see #2770)
Simon Marlow [Mon, 27 Apr 2009 12:47:11 +0000 (12:47 +0000)]
update build order comment
Manuel M T Chakravarty [Mon, 27 Apr 2009 14:03:16 +0000 (14:03 +0000)]
Equality constraint solver is now externally pure
- This patch changes the equality constraint solver such that it does not
instantiate any type variables that occur in the constraints that are to be
solved (or in the environment). Instead, it returns a bag of type bindings.
- If these type bindings (together with the other results of the solver) are
discarded, solver invocation has no effect (outside the solver) and can be
repeated (that's imported for TcSimplifyRestricted).
- For the type bindings to take effect, the caller of the solver needs to
execute them.
- The solver will still instantiate type variables thet were created during
solving (e.g., skolem flexibles used during type flattening).
See also http://hackage.haskell.org/trac/ghc/wiki/TypeFunctionsSolving
Ian Lynagh [Sun, 26 Apr 2009 16:02:04 +0000 (16:02 +0000)]
Improve the error message when we find a module in 2 places; trac #3183
Ian Lynagh [Sun, 26 Apr 2009 15:19:42 +0000 (15:19 +0000)]
Remove a comment (a bug to close, that has now been closed)
Ian Lynagh [Sun, 26 Apr 2009 11:42:15 +0000 (11:42 +0000)]
GHC new build system megapatch
Ben.Lippmeier@anu.edu.au [Sat, 25 Apr 2009 08:33:45 +0000 (08:33 +0000)]
Fix SPARC build, missing #include
Ian Lynagh [Fri, 24 Apr 2009 17:33:13 +0000 (17:33 +0000)]
Fix a lint failure when we have a ! (# ... #) pattern in a where clause
This showed up when converting ds057 to follow the new bang pattern rules,
in #2806.
Ian Lynagh [Fri, 24 Apr 2009 15:05:19 +0000 (15:05 +0000)]
Do the second part of #2806: Disallow unlifted types in ~ patterns
Ian Lynagh [Fri, 24 Apr 2009 14:02:27 +0000 (14:02 +0000)]
Tiny refactor
Ian Lynagh [Fri, 24 Apr 2009 12:47:54 +0000 (12:47 +0000)]
Require a bang pattern when unlifted types are where/let bound; #3182
For now we only get a warning, rather than an error, because the alex
and happy templates don't follow the new rules yet.
simonpj@microsoft.com [Fri, 24 Apr 2009 12:32:38 +0000 (12:32 +0000)]
Don't short out top-level indirections if there's a INLINE/NOINLINE pragma
The top-level indirection-shorting, done in SimplCore.shortOutIndirections,
was accidentally nuking a NOINLINE pragma. This patch adopts a slightly
more conservative approach to indirection-shorting, only doing so if
we do not thereby overwrite or lose some user-supplied pragmas.
Ben.Lippmeier@anu.edu.au [Fri, 24 Apr 2009 06:50:53 +0000 (06:50 +0000)]
Fix #3182: 64 bit FP value returned by C calls was getting trashed
movss instruction was being used to move the result into the
destination register instead of movsd
Simon Marlow [Fri, 24 Apr 2009 09:16:10 +0000 (09:16 +0000)]
add missing files (part of #3171 fix)
simonpj@microsoft.com [Fri, 24 Apr 2009 07:47:33 +0000 (07:47 +0000)]
Tiny optimisation to mkInlineMe
Simon Marlow [Thu, 23 Apr 2009 14:19:40 +0000 (14:19 +0000)]
Add EVENT_CREATE_SPARK_THREAD to replace EVENT_SPARK_TO_THREAD
Also some tidyups and renaming
Simon Marlow [Thu, 23 Apr 2009 11:30:02 +0000 (11:30 +0000)]
add getOrSetSignalHandlerStore, much like getOrSetTypeableStore
Part of the fix for #3171
Simon Marlow [Thu, 23 Apr 2009 11:29:39 +0000 (11:29 +0000)]
don't overwrite io_manager_pipe if it is already set
Part of the fix for #3171
simonpj@microsoft.com [Thu, 23 Apr 2009 09:44:01 +0000 (09:44 +0000)]
Simplify the placeholder binding for naughty record selectors
simonpj@microsoft.com [Thu, 23 Apr 2009 09:42:37 +0000 (09:42 +0000)]
Fix Trac #3176: intercept naughty record selectors
When making record selectors into ordinary functions (rather than
magial implicit Ids) I forgot that they could therefore show up in
the *local* TcEnv, not the global one. This fixes that problem,
and thereby Trac #3176.
simonpj@microsoft.com [Thu, 23 Apr 2009 09:39:45 +0000 (09:39 +0000)]
Better pretty-printing for IdDetails
Simon Marlow [Tue, 21 Apr 2009 13:56:24 +0000 (13:56 +0000)]
FIX #3166: include the fixity of classes and type synonyms in their fingerprints
Simon Marlow [Tue, 21 Apr 2009 13:10:38 +0000 (13:10 +0000)]
FIX #2682: banish silly cases of the "module Foo is not loaded" error
In GHCi if you say 'import Foo' meaning to load a package module Foo,
and Foo.hs is found on the search path, then GHCi replies "module Foo
is not loaded", because it knows Foo refers to the source file rather
than the package module, and you haven't loaded that module with
:load.
This is consistent with the usual module-finding semantics. However,
it isn't particularly useful. And it leads to silly problems like not
being able to start GHCi when you happen to be sitting in
libraries/base, because GHCi thinks the Prelude hasn't been loaded.
So now I've made a slight change to the way that 'import M' works: if
M is loaded, then it refers to the loaded module, otherwise it looks
for a package module M. This does what the reporter of #2682 wanted,
and since it turns an error condition into meaningful behaviour it
can't break anything.
The only undesirable consequence is that 'import M' might refer to a
different M than ':load M'. Hopefully that won't lead to confusion.
Simon Marlow [Tue, 21 Apr 2009 10:00:39 +0000 (10:00 +0000)]
FIX #2500: Don't log warnings in getHeader
Simon Marlow [Tue, 21 Apr 2009 09:46:06 +0000 (09:46 +0000)]
add a DO NOT EDIT comment
Simon Marlow [Tue, 21 Apr 2009 09:45:58 +0000 (09:45 +0000)]
FIX #3153: we lost an EOF sentinel in the event of a lexical error
Simon Marlow [Mon, 20 Apr 2009 14:25:25 +0000 (14:25 +0000)]
FIX #2845: Allow breakpoints on expressions with unlifted type
It turns out we can easily support breakpoints on expressions with
unlifted types, by translating
case tick# of _ -> e
into
let f = \s . case tick# of _ -> e
in f realWorld#
instead of just a plain let-binding. This is the same trick that GHC
uses for abstracting join points of unlifted type.
In #2845, GHC has eta-expanded the tick expression, changing the
result type from IO a to (# State#, a #), which was the reason the
tick was suddenly being ignored. By supporting ticks on unlifted
expressions we can make it work again, although some confusion might
arise because _result will no longer be available (it now has
unboxed-tuple type, so we can't bind it in the environment). The
underlying problem here is that GHC does transformations like
eta-expanding the tick expressions, and there's nothing we can do to
prevent that.
Simon Marlow [Fri, 3 Apr 2009 13:27:31 +0000 (13:27 +0000)]
remove now-unused atomic_modify_mutvar_mutex
Ben.Lippmeier@anu.edu.au [Mon, 20 Apr 2009 07:43:35 +0000 (07:43 +0000)]
SPARC NCG: Add a comment explaining why we can't used a pinned reg for gct
Can't use windowed regs because the window moves during a function
call. Can't use the global regs because they're reserved for other purposes.
Ben.Lippmeier@anu.edu.au [Mon, 20 Apr 2009 03:21:20 +0000 (03:21 +0000)]
SPARC NCG: validate fixes
Ben.Lippmeier@anu.edu.au [Mon, 20 Apr 2009 02:07:00 +0000 (02:07 +0000)]
SPARC NCG: Base freeRegs on includes/MachRegs.h again
pepe iborra [Sun, 19 Apr 2009 14:22:41 +0000 (14:22 +0000)]
Really fix Trac #2611 this time
My previous patch didn't completely solve the problem.
I believe I got it right this time.
pepe iborra [Sat, 18 Apr 2009 21:41:15 +0000 (21:41 +0000)]
Fix Trac #2611
Fix a bug in :print affecting data types with unboxed components
Ben.Lippmeier@anu.edu.au [Sat, 18 Apr 2009 08:22:53 +0000 (08:22 +0000)]
Fix #3170: Allow coalescing of the same node during register allocation.
The x86 code generator sometimes generates nop moves like
movl %vI_a %vI_a, but we'll just ignore them for coalescing
purposes instead of emitting an error. It shouldn't hurt anything.
David Waern [Sat, 18 Apr 2009 08:29:27 +0000 (08:29 +0000)]
Fix my previous patch about type parsing
I forgot to record some additional changes.
David Waern [Fri, 17 Apr 2009 14:58:55 +0000 (14:58 +0000)]
Simplify the type grammar
Simon P-J suggested the following simplifications in #3097:
* Allow nested foralls in `ctype` just like in `ctypedoc`.
* Use `gentype` rather than `type` in the LHS of type declarations.
* Inline `type` in `ctype`.
* Rename `gentype` to `type`.
This patch does this. Also, the equivalent thing is done for documented types.
donnie@darthik.com [Mon, 13 Apr 2009 01:17:45 +0000 (01:17 +0000)]
Removed unused function postEventTypeID.
postEventTypeID was used to post event marker IDs, but instead we just
call postWord16 directly; thus, this function is unused.
donnie@darthik.com [Mon, 13 Apr 2009 01:17:01 +0000 (01:17 +0000)]
For consistency, changed access of thread id to be through capability instead of directly from StgRegTable.
donnie@darthik.com [Mon, 13 Apr 2009 01:14:44 +0000 (01:14 +0000)]
Added new EventLog event: Spark to Thread.
donnie@darthik.com [Mon, 13 Apr 2009 01:11:40 +0000 (01:11 +0000)]
Changed postEvent last argument from "nat from" to "StgWord64 other".
StgWord64 other better represents this argument since it can be used as
any particular data, thus "other" and not the "from" capability as it was
previously strictly used. Also, StgWord64 is normally larger than type
nat to allow for larger data to be passed through the "other" argument.
donnie@darthik.com [Mon, 13 Apr 2009 01:07:05 +0000 (01:07 +0000)]
Fixed ThreadID to be defined as StgThreadID, not StgWord64. Changed CapabilityNum to CapNo. Added helper functions postCapNo() and postThreadID().
ThreadID was StgWord64, but should have been StgThreadID, which is
currently StgWord32. Changed name from CapabilityNum to CapNo to better
reflect naming in Capability struct where "no" is the capability number.
Modified EventLog.c to use the helper functions postCapNo() and
postThreadID () for CapNo and ThreadID.
donnie@darthik.com [Fri, 3 Apr 2009 03:43:22 +0000 (03:43 +0000)]
Fixed error in order of EventTypeDescriptions -- seq and par GC come after start and finish GC.
EventTypeDescriptions order must be synchronized with the event type
definitions in includes/EventLogFormat.h for the definitions to correctly
index with the matching description.
donnie@darthik.com [Fri, 3 Apr 2009 03:30:25 +0000 (03:30 +0000)]
Eventlog support for new event type: create spark.
Ian Lynagh [Tue, 7 Apr 2009 14:30:57 +0000 (14:30 +0000)]
Tweak a Show instance
simonpj@microsoft.com [Thu, 9 Apr 2009 14:40:04 +0000 (14:40 +0000)]
Fix Trac #3155: better error message when -XRankNTypes is omitted
This patch sligtly re-adjusts the way in which the syntax of types
is handled:
* In the lexer, '.' and '*' are always accepted in types
(previously it was conditional). This things can't mean
anything else in H98, which is the only reason for doing things
conditionally in the lexer.
* As a result '.' in types is never treated as an operator.
Instead, lacking a 'forall' keyword, it turns into a plain parse error.
* Test for -XKindSignatures in the renamer when processing
a) type variable bindings
b) types with sigs (ty :: kind-sig)
* Make -XKindSignatures be implied by -XTypeFamilies
Previously this was buried in the conditonal lexing of '*'
simonpj@microsoft.com [Thu, 9 Apr 2009 14:09:59 +0000 (14:09 +0000)]
Use return instead of returnM, and similar tidy-ups
Ian Lynagh [Sat, 4 Apr 2009 00:34:45 +0000 (00:34 +0000)]
Don't use thread local storage on x86/not-Linux
With the
On x86, use thread-local storage instead of stealing a reg for gct
patch, on Windows and OS X:
error: thread-local storage not supported for this target
Ian Lynagh [Tue, 31 Mar 2009 18:19:48 +0000 (18:19 +0000)]
Make some showSDoc's use OneLineMode rather than PageMode
Ian Lynagh [Tue, 31 Mar 2009 17:41:03 +0000 (17:41 +0000)]
Make the showSDoc definition more explicit
dias@eecs.tufts.edu [Fri, 3 Apr 2009 20:34:29 +0000 (20:34 +0000)]
eliminate warnings
dias@eecs.tufts.edu [Fri, 3 Apr 2009 20:15:04 +0000 (20:15 +0000)]
Debugging by Sesame Street:
One of these things is not like the others:
stdPattern :: [LRep] -> Maybe StgHalfWord
stdPattern reps
= case reps of
[] -> Just ARG_NONE -- just void args, probably
[N] -> Just ARG_N
[P] -> Just ARG_N
[F] -> Just ARG_F
[D] -> Just ARG_D
[L] -> Just ARG_L
Today's debugging session was brought to you by the letter P.
dias@eecs.tufts.edu [Tue, 31 Mar 2009 14:46:39 +0000 (14:46 +0000)]
Buggy optimizations caused function-call return to share the function's entry point
- Block concat and branch-chain elimination were allowing a function call
to return to the caller's entry point. But that doesn't leave anywhere
for the infotable on the stack, since the SP on return must be the same
as the SP on entry to the procedure.
dias@eecs.tufts.edu [Wed, 25 Mar 2009 16:38:15 +0000 (16:38 +0000)]
Better handling of node parameter in calling conventions
- Previously, the node was taken as a parameter, then ignored,
for static closures. Goofy. Now, the vestigial node parameters
are gone.
dias@eecs.tufts.edu [Mon, 23 Mar 2009 20:47:44 +0000 (20:47 +0000)]
When calling gc, avoid saving node in static closures
dias@eecs.tufts.edu [Mon, 23 Mar 2009 20:11:40 +0000 (20:11 +0000)]
Code simplifications due to call/return separation; some improvements to how node argument is managed
dias@eecs.tufts.edu [Mon, 23 Mar 2009 18:22:14 +0000 (18:22 +0000)]
Code simplification due to separate call/return conventions
dias@eecs.tufts.edu [Mon, 23 Mar 2009 17:47:00 +0000 (17:47 +0000)]
Calls with and without passing node arguments more clearly separated
dias@eecs.tufts.edu [Mon, 23 Mar 2009 17:28:37 +0000 (17:28 +0000)]
Another small step: call and return conventions specified separately when making calls
dias@eecs.tufts.edu [Mon, 23 Mar 2009 17:07:06 +0000 (17:07 +0000)]
Small step toward call-conv improvement: separate out calls and returns
Simon Marlow [Fri, 3 Apr 2009 12:18:16 +0000 (12:18 +0000)]
On x86, use thread-local storage instead of stealing a reg for gct
Benchmarks show that using TLS instead of stealing a register is
better by a few percent on x86, due to the lack of registers.
This only affects -threaded; without -threaded we're (now) using
static storage for the GC data.
Simon Marlow [Fri, 3 Apr 2009 12:14:43 +0000 (12:14 +0000)]
in the non-threaded RTS, use a static gc_thread structure
Simon Marlow [Fri, 3 Apr 2009 08:37:08 +0000 (08:37 +0000)]
small GC optimisation
Simon Marlow [Thu, 2 Apr 2009 10:56:43 +0000 (10:56 +0000)]
add -funroll-loops when compiling the parallel version of Evac.c too
Simon Marlow [Thu, 2 Apr 2009 10:55:40 +0000 (10:55 +0000)]
PACKAGE_CPP_OPTS += -DPAPI_INCLUDE_DIR=""
unbreaks ghc-pkg check in an installed GHC
simonpj@microsoft.com [Fri, 3 Apr 2009 08:46:34 +0000 (08:46 +0000)]
Adjust inlining heursitics
This patch is the result of a long series of nofib-based experiments
to improve GHC's inlining heuristics.
In the end, I'm not sure how worthwhile it all was: I only got a
1% decrease in code size
1% decrease in allocation
and I don't trust the runtime statistics enough to quote.
Still, in doing all this I tidied up the code quite a bit, and
I understand it much better now, so I'm going to commit it.
The main changes are in CoreUnfold, which has lots of new comments.
Other changes:
- litSize moves from Literal to CoreUnfold
- interestingArg moves from SimplUtils to CoreUnfold
- the default unfolding threshold (in StaticFlags)
reduces from 8 to 6 (since the size calculation
has changed a bit)
simonpj@microsoft.com [Fri, 3 Apr 2009 08:43:33 +0000 (08:43 +0000)]
Worker/wrapper should make INLINE if it doesn't w/w
If worker/wrapper decides not to w/w something on the grounds that
it's too small, it should add an INLINE pragma. Otherwise, later
in the day that small thing might now be big, and we'd wish we'd
done the w/w after all. This only made a difference in one nofib
program (bspt), but it's an easy change.
See Note [Don't w/w inline things (a) and (b)]
simonpj@microsoft.com [Fri, 3 Apr 2009 08:36:55 +0000 (08:36 +0000)]
Rewrite a good chunk of CoreArity
I found a couple of shortcomings in arity computation, and did
quite a bit of refactoring as a result. Regrettably, I have
forgotten the details, but I do remember that one part was to
do with the infamous "state hack". If we're going to use the
state-hack at all, we'd better do it right.
Anyway I think this is an improvement. The comments are more
up to date too, and more voluminous.
simonpj@microsoft.com [Thu, 2 Apr 2009 16:52:26 +0000 (16:52 +0000)]
Comments only; record remarks about removing 'type' nonterminal
simonpj@microsoft.com [Thu, 2 Apr 2009 16:51:01 +0000 (16:51 +0000)]
Don't float case expressions in full laziness
See Note [Case MFEs]; don't float case expressions from
a strict context.
simonpj@microsoft.com [Thu, 2 Apr 2009 15:28:34 +0000 (15:28 +0000)]
Fix Trac #3118: missing alternative
This patch fixes a rather obscure bug, whereby it's possible
for (case C a b of <alts>) to have altenatives that do not inclue
(C a b)! See Note [Unreachable code] in CoreUtils.
simonpj@microsoft.com [Thu, 2 Apr 2009 15:22:46 +0000 (15:22 +0000)]
Use a local interestingDict function instead of importing SimplUtils.interestingArg
I'm changing the details of SimplUtils.interstingArg, and don't want to
mess up the way Specialise works, so this patch makes a specilialised
(ha) function, Specialise.interestingDict, that is used locally.
simonpj@microsoft.com [Thu, 2 Apr 2009 15:18:46 +0000 (15:18 +0000)]
Better panic reporting
Make idDetails and idInfo into proper functions, rather than record
fields, so that we can report more informatively if you use thenm on
a non-Id.
simonpj@microsoft.com [Thu, 2 Apr 2009 15:16:16 +0000 (15:16 +0000)]
Fix pragmas in binder-swapping
In the case binder-swap operation (which is done in OccurAnal),
we deliberately use shadowing. But the new shadowing binder should
not inherit any fancy INLINE or NOINLINE pragmas from its parent.
simonpj@microsoft.com [Thu, 2 Apr 2009 15:14:06 +0000 (15:14 +0000)]
Float constants to top-level even in first full laziness pass
This patch changes the compilation pipeline flags so that constants
are floated to the top level even in the first full laziness pass.
For some reason this was not done before. Fixing this makes a big
improvement in 'spectral/rewrite', and has zero effect on any other
nofib benchmark.
Simon Marlow [Wed, 1 Apr 2009 10:52:56 +0000 (10:52 +0000)]
update the intro section
Simon Marlow [Wed, 1 Apr 2009 10:34:18 +0000 (10:34 +0000)]
Remove the "Installing GHC" section
1. it was out of date in various ways
2. this is not the place people look for installation instructions
3. we have installation instructions elsewhere (e.g. the INSTALL
file in a binary distribution)
4. the section "layout of installed files" is now on the wiki under
Building/Installing.
Ben.Lippmeier@anu.edu.au [Tue, 31 Mar 2009 04:51:45 +0000 (04:51 +0000)]
SPARC: Fix ffi019 split load/store of HsInt64 into two parts to respect alignment constraints
Ben.Lippmeier@anu.edu.au [Mon, 30 Mar 2009 05:17:32 +0000 (05:17 +0000)]
SPARC NCG: Fix 64 bit addition, carry bit wasn't getting set.
Ben.Lippmeier@anu.edu.au [Mon, 30 Mar 2009 05:03:46 +0000 (05:03 +0000)]
SPARC NCG: When getting a 64 bit word, promote halves to 64 bit before shifting
Ben.Lippmeier@anu.edu.au [Tue, 31 Mar 2009 05:19:00 +0000 (05:19 +0000)]
SPARC NCG: HpLim is now always stored on the stack, not in a register
This fixes the out of memory errors we were getting on sparc
after the following patch:
Fri Mar 13 03:45:16 PDT 2009 Simon Marlow <marlowsd@gmail.com>
* Instead of a separate context-switch flag, set HpLim to zero
Ignore-this:
6c5bbe1ce2c5ef551efe98f288483b0
This reduces the latency between a context-switch being triggered and
the thread returning to the scheduler, which in turn should reduce the
cost of the GC barrier when there are many cores.
David Waern [Tue, 31 Mar 2009 21:23:06 +0000 (21:23 +0000)]
Allow Haddock comments in type synonyms
We now use `ctypedoc` instead of `ctype` for type synonyms. `ctypedoc` was
previously only used for top-level type signatures. This change means that type
synonyms now can contain comments, just like top-level type signatures.
Note:
* I've modified `ctypedoc` so it allows implicit parameters and equational
constraints, just like ctype.
* Since `ctypedoc` allows nested foralls, we now allow that in type synonyms.
* I have inlined some productions into gentypedoc so that there is now a
non-doc version of every production with a 'doc' suffix. (Stylistic change
only, which should make the code easier to follow).
* It would have been nice to simplify the grammar by unifying `ctype` and
ctypedoc` into one production, allowing comments on types everywhere (and
rejecting them after parsing, where necessary). This is however not possible
since it leads to ambiguity. The reason is the support for comments on record
fields:
> data R = R { field :: Int -- ^ comment on the field }
If we allow comments on types here, it's not clear if the comment applies
to 'field' or to 'Int'. So we must use `ctype` to describe the type.
Ian Lynagh [Tue, 31 Mar 2009 13:40:58 +0000 (13:40 +0000)]
mkErrorAppDs now takes an SDoc rather than a String
This avoids some showSDoc's where the String then gets converted back
into an SDoc.