ghc-hetmet.git
16 years agoSimplify generation of PR dictionaries for products
Roman Leshchinskiy [Fri, 24 Aug 2007 07:19:25 +0000 (07:19 +0000)]
Simplify generation of PR dictionaries for products

16 years agoRemove unused vectorisation built-in
Roman Leshchinskiy [Fri, 24 Aug 2007 05:15:24 +0000 (05:15 +0000)]
Remove unused vectorisation built-in

16 years agoAdapt PArray instance generation to new scheme
Roman Leshchinskiy [Fri, 24 Aug 2007 05:12:42 +0000 (05:12 +0000)]
Adapt PArray instance generation to new scheme

16 years agoAdd UArr built-in
Roman Leshchinskiy [Fri, 24 Aug 2007 05:12:13 +0000 (05:12 +0000)]
Add UArr built-in

16 years agoModify generation of PR dictionaries for new scheme
Roman Leshchinskiy [Fri, 24 Aug 2007 04:31:44 +0000 (04:31 +0000)]
Modify generation of PR dictionaries for new scheme

16 years agoRefactoring
Roman Leshchinskiy [Fri, 24 Aug 2007 04:09:01 +0000 (04:09 +0000)]
Refactoring

16 years agoRemove dead code
Roman Leshchinskiy [Fri, 24 Aug 2007 03:57:51 +0000 (03:57 +0000)]
Remove dead code

16 years agoFix buildFromPRepr
Roman Leshchinskiy [Fri, 24 Aug 2007 03:57:00 +0000 (03:57 +0000)]
Fix buildFromPRepr

16 years agoMove code
Roman Leshchinskiy [Fri, 24 Aug 2007 03:29:30 +0000 (03:29 +0000)]
Move code

16 years agoMove code
Roman Leshchinskiy [Fri, 24 Aug 2007 03:27:43 +0000 (03:27 +0000)]
Move code

16 years agoDelete dead code
Roman Leshchinskiy [Fri, 24 Aug 2007 03:15:04 +0000 (03:15 +0000)]
Delete dead code

16 years agoChange buildToPRepr to work with the new representation scheme
Roman Leshchinskiy [Fri, 24 Aug 2007 03:14:07 +0000 (03:14 +0000)]
Change buildToPRepr to work with the new representation scheme

16 years agoRemove Embed and related stuff from vectorisation
Roman Leshchinskiy [Fri, 24 Aug 2007 02:30:30 +0000 (02:30 +0000)]
Remove Embed and related stuff from vectorisation

16 years agoEncode generic representation of vectorised TyCons by a data type
Roman Leshchinskiy [Fri, 24 Aug 2007 01:21:40 +0000 (01:21 +0000)]
Encode generic representation of vectorised TyCons by a data type

16 years agoRemove dead code
Roman Leshchinskiy [Thu, 23 Aug 2007 13:58:10 +0000 (13:58 +0000)]
Remove dead code

16 years agoConversions to/from generic array representation (not finished yet)
Roman Leshchinskiy [Thu, 23 Aug 2007 13:56:49 +0000 (13:56 +0000)]
Conversions to/from generic array representation (not finished yet)

16 years agoUse n-ary sums and products for NDP's generic representation
Roman Leshchinskiy [Thu, 23 Aug 2007 06:09:45 +0000 (06:09 +0000)]
Use n-ary sums and products for NDP's generic representation

Originally, we wanted to only use binary ones, at least initially. But this
would a lot of fiddling with selectors when converting to/from generic
array representations. This is both inefficient and hard to implement.
Instead, we will limit the arity of our sums/product representation to, say,
16 (it's 3 at the moment) and initially refuse to vectorise programs for which
this is not sufficient. This allows us to implement everything in the library.
Later, we can implement the necessary splitting.

16 years agoFix where all the documentation gets installed
Ian Lynagh [Thu, 30 Aug 2007 22:37:40 +0000 (22:37 +0000)]
Fix where all the documentation gets installed
The paths can also now be overridden with the standard configure flags
--docdir=, --htmldir= etc. We were always advertising these, but now we
actually obey them.

16 years agoAdded decidability check for type instances
Manuel M T Chakravarty [Thu, 30 Aug 2007 14:49:01 +0000 (14:49 +0000)]
Added decidability check for type instances

16 years agoWarning police
Pepe Iborra [Wed, 29 Aug 2007 18:31:55 +0000 (18:31 +0000)]
Warning police

16 years agoUse a Data.Sequence instead of a list in cvReconstructType
Pepe Iborra [Wed, 29 Aug 2007 17:51:19 +0000 (17:51 +0000)]
Use a Data.Sequence instead of a list in cvReconstructType

While I was there I removed some trailing white space

16 years agoFix a bug in RtClosureInspect.cvReconstructType.
Pepe Iborra [Wed, 29 Aug 2007 17:48:42 +0000 (17:48 +0000)]
Fix a bug in RtClosureInspect.cvReconstructType.
Test is print025

16 years agoWarning police
Pepe Iborra [Wed, 29 Aug 2007 16:56:53 +0000 (16:56 +0000)]
Warning police

16 years agoUNDO: Extend ModBreaks with the srcspan's of the enclosing expressions
Pepe Iborra [Wed, 29 Aug 2007 10:23:14 +0000 (10:23 +0000)]
UNDO: Extend ModBreaks with the srcspan's of the enclosing expressions

Remnants of :stepover

16 years agoremove "special Ids" section, replace with a link to GHC.Prim
Simon Marlow [Thu, 30 Aug 2007 11:21:39 +0000 (11:21 +0000)]
remove "special Ids" section, replace with a link to GHC.Prim
This documentation was just duplicating what is in GHC.Prim now.

16 years agoexpand docs for unsafeCoerce#, as a result of investigations for #1616
Simon Marlow [Thu, 30 Aug 2007 11:19:09 +0000 (11:19 +0000)]
expand docs for unsafeCoerce#, as a result of investigations for #1616

16 years agoRemove text about ghcprof. It almost certainly doesn't work.
Simon Marlow [Wed, 29 Aug 2007 12:21:26 +0000 (12:21 +0000)]
Remove text about ghcprof.  It almost certainly doesn't work.

16 years agofix compiling GHC 6.7+ with itself - compat needs -package containers now
Simon Marlow [Wed, 29 Aug 2007 11:35:00 +0000 (11:35 +0000)]
fix compiling GHC 6.7+ with itself - compat needs -package containers now

16 years agofix typo
Simon Marlow [Fri, 24 Aug 2007 14:10:39 +0000 (14:10 +0000)]
fix typo

16 years agono -auto-all for CorePrep
Simon Marlow [Wed, 29 Aug 2007 09:24:14 +0000 (09:24 +0000)]
no -auto-all for CorePrep

16 years agoimprovements to findPtr(), a useful hack for space-leak debugging in gdb
Simon Marlow [Wed, 29 Aug 2007 09:24:00 +0000 (09:24 +0000)]
improvements to findPtr(), a useful hack for space-leak debugging in gdb

16 years agofix up some old text, remove things that aren't true any more
Simon Marlow [Tue, 28 Aug 2007 12:58:21 +0000 (12:58 +0000)]
fix up some old text, remove things that aren't true any more

16 years agoWindows: remove the {Enter,Leave}CricialSection wrappers
Simon Marlow [Wed, 29 Aug 2007 10:48:11 +0000 (10:48 +0000)]
Windows: remove the {Enter,Leave}CricialSection wrappers
The C-- parser was missing the "stdcall" calling convention for
foreign calls, but once added we can call {Enter,Leave}CricialSection
directly.

16 years agoWibble
Pepe Iborra [Wed, 29 Aug 2007 08:53:05 +0000 (08:53 +0000)]
Wibble

16 years agoFIX: Remove accidential change to darcs-all in type families patch
Manuel M T Chakravarty [Wed, 29 Aug 2007 01:00:11 +0000 (01:00 +0000)]
FIX: Remove accidential change to darcs-all in type families patch
- The type families patch includes a change to darcs-all that breaks it for
  ssh repos at least for Perl 5.8.8 (on MacOS).
- My Perl-fu is not sufficient to try to fix the modification, which was
  supposed to improve darcs-all on windows, so I just revert to the old
  code.

16 years agoRemove INSTALL_INCLUDES; no longer used
Ian Lynagh [Tue, 28 Aug 2007 20:56:36 +0000 (20:56 +0000)]
Remove INSTALL_INCLUDES; no longer used

16 years agoUse DESTDIR when installing
Ian Lynagh [Tue, 28 Aug 2007 20:51:19 +0000 (20:51 +0000)]
Use DESTDIR when installing

16 years agoCopy LICENSE files into the bindist, as Cabal now installs them
Ian Lynagh [Tue, 28 Aug 2007 13:04:28 +0000 (13:04 +0000)]
Copy LICENSE files into the bindist, as Cabal now installs them

16 years agoBetter handling of join points in spill cleaner 2007-08-28
Ben.Lippmeier@anu.edu.au [Tue, 28 Aug 2007 13:30:50 +0000 (13:30 +0000)]
Better handling of join points in spill cleaner

16 years agocomment wibble
Ben.Lippmeier@anu.edu.au [Fri, 24 Aug 2007 16:03:14 +0000 (16:03 +0000)]
comment wibble

16 years agoAdd count of reg-reg moves remaining for linear allocator stats
Ben.Lippmeier@anu.edu.au [Fri, 24 Aug 2007 15:57:32 +0000 (15:57 +0000)]
Add count of reg-reg moves remaining for linear allocator stats

16 years agoAdd a count of how many spill/reloads/reg-reg-moves remain to dump-asm-stats
Ben.Lippmeier@anu.edu.au [Fri, 24 Aug 2007 15:21:42 +0000 (15:21 +0000)]
Add a count of how many spill/reloads/reg-reg-moves remain to dump-asm-stats

16 years agoUse INSTALL_HEADERS in includes/ rather than abusing INSTALL_DATAS
Ian Lynagh [Tue, 28 Aug 2007 12:49:18 +0000 (12:49 +0000)]
Use INSTALL_HEADERS in includes/ rather than abusing INSTALL_DATAS

16 years agoMake some header-installing machinery, and use it to install gmp.h
Ian Lynagh [Tue, 28 Aug 2007 12:47:59 +0000 (12:47 +0000)]
Make some header-installing machinery, and use it to install gmp.h

16 years agoTidy up the gmp/ Makefile a bit
Ian Lynagh [Tue, 28 Aug 2007 12:32:18 +0000 (12:32 +0000)]
Tidy up the gmp/ Makefile a bit

16 years ago:stepover ---> :steplocal, :stepmodule
Pepe Iborra [Mon, 27 Aug 2007 18:09:49 +0000 (18:09 +0000)]
:stepover ---> :steplocal, :stepmodule

 :stepover is declared a failed experiment.
:steplocal steps only on ticks contained in the current
top level declaration.
:stepmodule steps only on ticks contained on the current
 module.
The current top level declaration and module are with
 respect to the breakpoint we are stopped on.

 The main reason for the failure of :stepover
(apart from lacking a lexical call stack of course)
is that it fails to detect when the expression being
evaluated is "complete", i.e. there are no ticks left in it.
My assumption of the rightmost tick as the "last one",
signaling that the expression is completely evaluated,
is not true at all under laziness.
This assumption was key in the implementation of :stepover.

16 years agoUse a version of obtainTerm that takes a max depth bound
Pepe Iborra [Mon, 27 Aug 2007 17:23:15 +0000 (17:23 +0000)]
Use a version of obtainTerm that takes a max depth bound
when printing the contents of binding at a breakpoint

16 years agoBe more careful when calculating the enclosing top level declaration of a srcspan
Pepe Iborra [Mon, 27 Aug 2007 11:44:37 +0000 (11:44 +0000)]
Be more careful when calculating the enclosing top level declaration of a srcspan

16 years agoStyle: remove trailing spaces
Pepe Iborra [Sun, 26 Aug 2007 21:33:57 +0000 (21:33 +0000)]
Style: remove trailing spaces

16 years agoPrint contents of bindings when stopping at a breakpoint
Pepe Iborra [Sun, 26 Aug 2007 21:33:39 +0000 (21:33 +0000)]
Print contents of bindings when stopping at a breakpoint

16 years agoA partial attempt to improve :stepover
Pepe Iborra [Fri, 24 Aug 2007 13:40:22 +0000 (13:40 +0000)]
A partial attempt to improve :stepover

  With this patch, :stepover can effectively appear to step over recursive calls and
  calls to locally bound functions (in a where clause).

  However, when we run out of ticks in the current expression,
  the illusion vanishes and laziness brings us to the body of the last function
  we "stepped over".
  This is not desired at all, it is potentially very confusing.
  As a countermeasure, when this happens :stepover emits a warning

     "Warning: no more breakpoints in this function body, switching to :step"

16 years agoBetter document :stepover and its limitations
Pepe Iborra [Wed, 22 Aug 2007 08:42:54 +0000 (08:42 +0000)]
Better document :stepover and its limitations

:stepover only works lexically locally, in the context of the
current expression. I have tried to make this point clear
in the users guide with an example.

16 years agoFIX #1533: foreign exporing the same identifier multiple times gave a link error
Simon Marlow [Tue, 28 Aug 2007 09:15:50 +0000 (09:15 +0000)]
FIX #1533: foreign exporing the same identifier multiple times gave a link error
We were generating a new top-level binding derived from the name of
the existing top-level name, and making the name external.  Multiple
instances therefore clashed.  The fix is to make each name unique, by
appending an actual Unique to the derived name.

16 years agoconvert to use System.FilePath
Simon Marlow [Sun, 26 Aug 2007 15:19:03 +0000 (15:19 +0000)]
convert to use System.FilePath

16 years agoRefactoring only: remove [Id] field from ForeignStubs
Simon Marlow [Sun, 26 Aug 2007 07:33:22 +0000 (07:33 +0000)]
Refactoring only: remove [Id] field from ForeignStubs
We used to pass the list of top-level foreign exported bindings to the
code generator so that it could create StablePtrs for them in the
stginit code.  Now we don't use stginit unless profiling, and the
StablePtrs are generated by C functions marked with
attribute((constructor)).  This patch removes various bits associated
with the old way of doing things, which were previously left in place
in case we wanted to switch back, I presume.  Also I refactored
dsForeigns to clean it up a bit.

16 years agoType checking for type synonym families
Manuel M T Chakravarty [Tue, 28 Aug 2007 06:18:51 +0000 (06:18 +0000)]
Type checking for type synonym families

This patch introduces type checking for type families of which associated
type synonyms are a special case. E.g.

        type family Sum n m

        type instance Sum Zero n = n
        type instance Sum (Succ n) m = Succ (Sum n m)

where

        data Zero       -- empty type
        data Succ n     -- empty type

In addition we support equational constraints of the form:

        ty1 ~ ty2

(where ty1 and ty2 are arbitrary tau types) in any context where
type class constraints are already allowed, e.g.

        data Equals a b where
                Equals :: a ~ b => Equals a b

The above two syntactical extensions are disabled by default. Enable
with the -XTypeFamilies flag.

For further documentation about the patch, see:

        * the master plan
          http://hackage.haskell.org/trac/ghc/wiki/TypeFunctions

        * the user-level documentation
          http://haskell.org/haskellwiki/GHC/Indexed_types

The patch is mostly backwards compatible, except for:

        * Some error messages have been changed slightly.

        * Type checking of GADTs now requires a bit more type declarations:
          not only should the type of a GADT case scrutinee be given, but also
          that of any identifiers used in the branches and the return type.

Please report any unexpected behavior and incomprehensible error message
for existing code.

Contributors (code and/or ideas):
        Tom Schrijvers
        Manuel Chakravarty
        Simon Peyton-Jones
        Martin Sulzmann
with special thanks to Roman Leshchinskiy

16 years agoFix retainer profiling Before_type_family_merge
Ian Lynagh [Sun, 26 Aug 2007 14:45:40 +0000 (14:45 +0000)]
Fix retainer profiling

16 years agoOn Windows, stage1 also deps on stamp.inplace-gcc-lib
Ian Lynagh [Sun, 26 Aug 2007 13:38:30 +0000 (13:38 +0000)]
On Windows, stage1 also deps on stamp.inplace-gcc-lib

16 years agoSynched GHC's .spec file with recent changes
sven.panne@aedion.de [Sun, 26 Aug 2007 09:30:26 +0000 (09:30 +0000)]
Synched GHC's .spec file with recent changes
Fixed hsc2hs handling and added hpc binary. Explicitly generate Cabal docs.
Note that the Haddock documentation is currently not in the right place,
this needs more investigation, but at least we can build a valid RPM now.

16 years agoCheck that exported modules were actually imported; fixes #1384
Ian Lynagh [Sun, 26 Aug 2007 00:12:32 +0000 (00:12 +0000)]
Check that exported modules were actually imported; fixes #1384

16 years agoBe more flexible when trying to make scripts executable
sven.panne@aedion.de [Sat, 25 Aug 2007 19:32:16 +0000 (19:32 +0000)]
Be more flexible when trying to make scripts executable

16 years agoDon't put push-all and darcs-all in source dists
Ian Lynagh [Sat, 25 Aug 2007 17:55:02 +0000 (17:55 +0000)]
Don't put push-all and darcs-all in source dists
We don't have the _darcs directories, so they won't work.

16 years agoAdd a missing "entryCode"
Ian Lynagh [Sat, 25 Aug 2007 17:12:31 +0000 (17:12 +0000)]
Add a missing "entryCode"
This was causing us to try to jump to the address of an infotable when
unregisterised, leading to a segfault.

16 years agoFix the IF_DEBUG(interpreter in StgCRun.c
Ian Lynagh [Sat, 25 Aug 2007 13:08:10 +0000 (13:08 +0000)]
Fix the IF_DEBUG(interpreter in StgCRun.c

16 years agoAdd "Support SMP" and "RTS ways" to the compiler --info output
Ian Lynagh [Sat, 25 Aug 2007 01:57:54 +0000 (01:57 +0000)]
Add "Support SMP" and "RTS ways" to the compiler --info output

16 years agoOverhaul boot script
sven.panne@aedion.de [Sat, 25 Aug 2007 13:30:52 +0000 (13:30 +0000)]
Overhaul boot script
* Call autoreconf only if necessary, avoiding a strict tool dependency.
* Be a good *nix citizen and don't clutter the output when everything is fine.
* Make a few more scripts executable.

16 years agoAdded a few missing files to a source distribution
sven.panne@aedion.de [Sat, 25 Aug 2007 12:11:50 +0000 (12:11 +0000)]
Added a few missing files to a source distribution

16 years agoFixed test for pwd executable
sven.panne@aedion.de [Sat, 25 Aug 2007 11:55:19 +0000 (11:55 +0000)]
Fixed test for pwd executable
Added a missing "test" and replaced the "-e" option of the shell-builtin
"test" with "-f". The former is not portable, see autoconf documentation.

16 years agoJump through some hoops to make the Windows SEH exception stuff happy
Ian Lynagh [Fri, 24 Aug 2007 22:23:17 +0000 (22:23 +0000)]
Jump through some hoops to make the Windows SEH exception stuff happy

16 years agoCheck haddocking works when validating
Ian Lynagh [Fri, 24 Aug 2007 19:56:09 +0000 (19:56 +0000)]
Check haddocking works when validating

16 years agoAdd "have interpreter" to the compiler --info output
Ian Lynagh [Fri, 24 Aug 2007 21:30:24 +0000 (21:30 +0000)]
Add "have interpreter" to the compiler --info output

16 years agoIn ghc-inplace, flush stdout after giving debugging output
Ian Lynagh [Fri, 24 Aug 2007 13:27:02 +0000 (13:27 +0000)]
In ghc-inplace, flush stdout after giving debugging output

16 years agoFollow Array changes (adding numElements field)
Ian Lynagh [Fri, 10 Aug 2007 14:16:14 +0000 (14:16 +0000)]
Follow Array changes (adding numElements field)

16 years agoCabal upstream API change -- installPackage to use installDirTemplates
Clemens Fruhwirth [Fri, 24 Aug 2007 18:13:25 +0000 (18:13 +0000)]
Cabal upstream API change -- installPackage to use installDirTemplates

16 years agoErase unneeded spill/reloads after register allocation
Ben.Lippmeier@anu.edu.au [Fri, 24 Aug 2007 14:31:32 +0000 (14:31 +0000)]
Erase unneeded spill/reloads after register allocation

16 years agoBe more paranoid about not creating space leaks in coloring allocator
Ben.Lippmeier@anu.edu.au [Fri, 24 Aug 2007 10:13:05 +0000 (10:13 +0000)]
Be more paranoid about not creating space leaks in coloring allocator

16 years agoShow spill/reload pseudo instrs in regalloc stage dump
Ben.Lippmeier@anu.edu.au [Fri, 24 Aug 2007 09:57:34 +0000 (09:57 +0000)]
Show spill/reload pseudo instrs in regalloc stage dump

16 years agowindres also needs the PATH workaround, because it runs gcc
Simon Marlow [Fri, 24 Aug 2007 14:33:30 +0000 (14:33 +0000)]
windres also needs the PATH workaround, because it runs gcc

16 years agofreeProfiling1() needs to be after endProfiling()
Simon Marlow [Fri, 24 Aug 2007 13:19:25 +0000 (13:19 +0000)]
freeProfiling1() needs to be after endProfiling()
Fixes another crash with biographical profiling.

16 years agofix getting the exit code
Simon Marlow [Fri, 24 Aug 2007 13:12:45 +0000 (13:12 +0000)]
fix getting the exit code

16 years agoFIX #1519, crash in biographical profiling
Simon Marlow [Fri, 24 Aug 2007 10:18:03 +0000 (10:18 +0000)]
FIX #1519, crash in biographical profiling
In eval_thunk_selector(), when filling the slop for a closure we've
just updated, we were using the wrong info table.  This bug appears to
have been in 6.6, too.

16 years agofix +RTS -V0 when not using -threaded
Simon Marlow [Fri, 24 Aug 2007 09:55:32 +0000 (09:55 +0000)]
fix +RTS -V0 when not using -threaded

16 years agoFix info table for stg_ap_1_upd, it stated 1 non-ptr, should be zero
Simon Marlow [Fri, 24 Aug 2007 09:54:40 +0000 (09:54 +0000)]
Fix info table for stg_ap_1_upd, it stated 1 non-ptr, should be zero
Fixes a crash I encountered with profiling and -debug

16 years agoclean doc-index-*.html (for older versions of Haddock)
Simon Marlow [Fri, 24 Aug 2007 08:24:06 +0000 (08:24 +0000)]
clean doc-index-*.html (for older versions of Haddock)

16 years agoadd pwd to SUBDIRS, so it gets cleaned
Simon Marlow [Fri, 24 Aug 2007 08:23:53 +0000 (08:23 +0000)]
add pwd to SUBDIRS, so it gets cleaned

16 years agoinclude bindisttest in source distributions (fixes srcdist)
Simon Marlow [Fri, 24 Aug 2007 08:08:47 +0000 (08:08 +0000)]
include bindisttest in source distributions (fixes srcdist)

16 years agoAdd spill/reload pseudo instrs to MachInstrs
Ben.Lippmeier@anu.edu.au [Thu, 23 Aug 2007 16:57:44 +0000 (16:57 +0000)]
Add spill/reload pseudo instrs to MachInstrs

Spiller can now insert spill/reload instrs without having to
worry about the current stack delta. Generation of actual machine
instructions for spills/reloads is deferred until after register
allocation proper.

16 years agoImprove docs for mdo
simonpj@microsoft.com [Fri, 24 Aug 2007 09:07:26 +0000 (09:07 +0000)]
Improve docs for mdo

16 years agoCorrect link
simonpj@microsoft.com [Fri, 24 Aug 2007 09:07:16 +0000 (09:07 +0000)]
Correct link

16 years agoFix typo in comment
Michael D. Adams [Mon, 6 Aug 2007 05:06:55 +0000 (05:06 +0000)]
Fix typo in comment

16 years agoAdded a case for non-zero arity INFO_TABLE_FUN to the Cmm parser.
Michael D. Adams [Mon, 6 Aug 2007 05:00:19 +0000 (05:00 +0000)]
Added a case for non-zero arity INFO_TABLE_FUN to the Cmm parser.
This is needed in order to write Main_main from pure Cmm.

16 years agoChange to the Cmm syntax. Put gc_block before update_frame.
Michael D. Adams [Mon, 6 Aug 2007 04:50:11 +0000 (04:50 +0000)]
Change to the Cmm syntax. Put gc_block before update_frame.
This order matches the symantic order better since
the GC check is done before the update frame is pushed.

16 years agoFix the calling convention order in the CPS
Michael D. Adams [Mon, 6 Aug 2007 04:48:11 +0000 (04:48 +0000)]
Fix the calling convention order in the CPS
It now matches the order used by the existing codeGen.

16 years agoRegalloc stage dump in right order
Ben.Lippmeier@anu.edu.au [Thu, 23 Aug 2007 13:34:07 +0000 (13:34 +0000)]
Regalloc stage dump in right order

16 years agoEliminate more dead reg->reg moves in linear allocator
Ben.Lippmeier@anu.edu.au [Thu, 23 Aug 2007 13:26:14 +0000 (13:26 +0000)]
Eliminate more dead reg->reg moves in linear allocator

16 years agoHandle DOA vregs when building spill stats
Ben.Lippmeier@anu.edu.au [Thu, 23 Aug 2007 12:10:09 +0000 (12:10 +0000)]
Handle DOA vregs when building spill stats

16 years agoUse UniqSet instead of Data.Set
Ben.Lippmeier@anu.edu.au [Thu, 23 Aug 2007 09:48:19 +0000 (09:48 +0000)]
Use UniqSet instead of Data.Set

16 years agocomment wibble
Ben.Lippmeier@anu.edu.au [Wed, 22 Aug 2007 17:11:28 +0000 (17:11 +0000)]
comment wibble

16 years agoRefactor cmmNativeGen so dumps can be emitted inline with NCG stages
Ben.Lippmeier@anu.edu.au [Wed, 22 Aug 2007 17:04:12 +0000 (17:04 +0000)]
Refactor cmmNativeGen so dumps can be emitted inline with NCG stages

16 years agoImprove handling of inline pragmas, esp where type applications are involved
simonpj@microsoft.com [Thu, 23 Aug 2007 15:20:27 +0000 (15:20 +0000)]
Improve handling of inline pragmas, esp where type applications are involved

See the definition of splitInlineCont for what this is about.
Cures Trac #1627.
Test is simpl017