packages/integer-gmp.git
9 years agoWhitespace only, in gmp-wrappers.cmm
Ian Lynagh [Tue, 21 Jul 2009 20:26:40 +0000 (20:26 +0000)] 
Whitespace only, in gmp-wrappers.cmm

9 years agoMake integer-gmp suitable to be used directly, rather than via integer
Ian Lynagh [Tue, 21 Jul 2009 18:53:02 +0000 (18:53 +0000)] 
Make integer-gmp suitable to be used directly, rather than via integer

9 years agoAdd GmpDerivedConstants.h dependencies for all ways, not just vanilla
Ian Lynagh [Sat, 18 Jul 2009 14:56:28 +0000 (14:56 +0000)] 
Add GmpDerivedConstants.h dependencies for all ways, not just vanilla

9 years agofix mk/build.mk for HaveLibGmp/HaveFrameworkGMP
Austin Seipp [Mon, 6 Jul 2009 21:39:37 +0000 (21:39 +0000)] 
fix mk/build.mk for HaveLibGmp/HaveFrameworkGMP
Previously if you were to set HaveFrameworkGMP/HaveLibGmp in
mk/build.mk (e.g. on mac os) the settings would be ignored, as
gmp/config.mk is included late in GHC's ghc.mk, meaning autoconf
overrides your settings. Now they ignore the values picked up by
configure if you set these variables.

9 years agoCheck for whether we have gmp.h, as well as whether we have the gmp library
Ian Lynagh [Thu, 25 Jun 2009 16:29:33 +0000 (16:29 +0000)] 
Check for whether we have gmp.h, as well as whether we have the gmp library

9 years agoFix the case of HaveFrameworkGMP values
Ian Lynagh [Thu, 25 Jun 2009 16:25:44 +0000 (16:25 +0000)] 
Fix the case of HaveFrameworkGMP values

9 years agofix the dependencies on GmpDerivedConstants.h when HaveLibGmp==YES 2009-06-25
Simon Marlow [Tue, 23 Jun 2009 14:09:22 +0000 (14:09 +0000)] 
fix the dependencies on GmpDerivedConstants.h when HaveLibGmp==YES

9 years agoMake this package now really integer-gmp, rather than pretending it's integer
Ian Lynagh [Mon, 22 Jun 2009 22:18:12 +0000 (22:18 +0000)] 
Make this package now really integer-gmp, rather than pretending it's integer

9 years agodon't include config.mk if we're cleaning
Simon Marlow [Tue, 23 Jun 2009 10:03:35 +0000 (10:03 +0000)] 
don't include config.mk if we're cleaning

9 years agoPatch GMP to always use the GHC allocation functions
Ian Lynagh [Sun, 21 Jun 2009 18:29:57 +0000 (18:29 +0000)] 
Patch GMP to always use the GHC allocation functions
This works around a crash (only on OS X for some reason?) where the GHCi
GMP uses the systems memory allocator. We should fix this properly, by
making ghci run constructor functions.

9 years agogmp build tweaks
Ian Lynagh [Wed, 17 Jun 2009 12:17:20 +0000 (12:17 +0000)] 
gmp build tweaks

9 years agoMake configure fail if deriving the constants fails
Ian Lynagh [Tue, 16 Jun 2009 18:30:05 +0000 (18:30 +0000)] 
Make configure fail if deriving the constants fails

9 years agoImprove the configure script
Ian Lynagh [Tue, 16 Jun 2009 17:37:53 +0000 (17:37 +0000)] 
Improve the configure script

9 years agoDon't put "extra-libraries: gmp" in the cabal file; it comes from the buildinfo file
Ian Lynagh [Mon, 15 Jun 2009 20:25:12 +0000 (20:25 +0000)] 
Don't put "extra-libraries: gmp" in the cabal file; it comes from the buildinfo file

9 years agoFixes for building on machines that don't have gmp
Ian Lynagh [Mon, 15 Jun 2009 20:23:52 +0000 (20:23 +0000)] 
Fixes for building on machines that don't have gmp

9 years agoMove the int64 conversion functions here, from ghc-prim
Ian Lynagh [Mon, 15 Jun 2009 20:17:32 +0000 (20:17 +0000)] 
Move the int64 conversion functions here, from ghc-prim

9 years agoMove gmp to here, from the GHC repo
Ian Lynagh [Sun, 14 Jun 2009 18:32:55 +0000 (18:32 +0000)] 
Move gmp to here, from the GHC repo

9 years agoAdd a configure script and rely on local definitions of derived constants
Duncan Coutts [Sat, 13 Jun 2009 14:46:10 +0000 (14:46 +0000)] 
Add a configure script and rely on local definitions of derived constants

9 years agoTweak the small integer case of gcdInteger for better optimisation
Duncan Coutts [Sat, 13 Jun 2009 13:40:49 +0000 (13:40 +0000)] 
Tweak the small integer case of gcdInteger for better optimisation
The gcdInt function in the base package now calls gcdInteger with
two small integers. With this weak, the optimiser generates a base
gcdInt that directly calls the gcdInt# primop from this package.
This means there should be no additional overhead compared to when
the base gcdInt called the gcdInt# primop directly.

9 years agoImplement the gmp primops in the integer-gmp package using cmm
Duncan Coutts [Sat, 13 Jun 2009 13:37:50 +0000 (13:37 +0000)] 
Implement the gmp primops in the integer-gmp package using cmm

9 years agoAdd __encodeDouble and __encodeFloat
Duncan Coutts [Thu, 11 Jun 2009 15:49:51 +0000 (15:49 +0000)] 
Add __encodeDouble and __encodeFloat
Moved here from the rts.

9 years agoRemove the unused decodeFloatInteger
Ian Lynagh [Tue, 2 Jun 2009 18:22:20 +0000 (18:22 +0000)] 
Remove the unused decodeFloatInteger

9 years agoUse a bang pattern when we where/let-bind values with unlifted types
Ian Lynagh [Fri, 24 Apr 2009 12:53:37 +0000 (12:53 +0000)] 
Use a bang pattern when we where/let-bind values with unlifted types

9 years agoDon't set -Wall -Werror in the .cabal file
Ian Lynagh [Thu, 12 Mar 2009 12:57:09 +0000 (12:57 +0000)] 
Don't set -Wall -Werror in the .cabal file

9 years agoDon't hide GHC.Integer from haddock; fixes trac #2839
Ian Lynagh [Wed, 11 Feb 2009 16:26:19 +0000 (16:26 +0000)] 
Don't hide GHC.Integer from haddock; fixes trac #2839
This works around haddock not exporting docs across package boundaries

9 years agoRequire Cabal version >= 1.6
Ian Lynagh [Thu, 22 Jan 2009 01:13:10 +0000 (01:13 +0000)] 
Require Cabal version >= 1.6

9 years agoAdd "bug-reports" and "source-repository" info to the Cabal file
Ian Lynagh [Wed, 21 Jan 2009 18:24:59 +0000 (18:24 +0000)] 
Add "bug-reports" and "source-repository" info to the Cabal file

9 years agoFix performance regression in quotRemInteger/divModInteger
Simon Marlow [Thu, 20 Nov 2008 09:58:15 +0000 (09:58 +0000)] 
Fix performance regression in quotRemInteger/divModInteger
This fixes most of the performance regression in these functions
between 6.8.3 and 6.10.1, it was due to a slight difference in the
strictness between the old and new versions (see comments for
details).  There's still a few percent loss in performance in the div
test in nofib/spectral/integer that I haven't tracked down as yet.

10 years agoMove the Integer type definition into GHC.Integer.Internals
Ian Lynagh [Thu, 25 Sep 2008 13:39:35 +0000 (13:39 +0000)] 
Move the Integer type definition into GHC.Integer.Internals
This means that we can export the constructors, but still keep the
GHC.Integer interface generic.

10 years agoPad version number to 0.1.0.0
Ian Lynagh [Sat, 20 Sep 2008 16:01:39 +0000 (16:01 +0000)] 
Pad version number to 0.1.0.0

10 years agoAdd a .darcs-boring file 6_10_branch_has_been_forked
Ian Lynagh [Fri, 27 Jun 2008 15:54:53 +0000 (15:54 +0000)] 
Add a .darcs-boring file

10 years agoAvoid using deprecated flags
Ian Lynagh [Mon, 16 Jun 2008 14:53:51 +0000 (14:53 +0000)] 
Avoid using deprecated flags

10 years agoWibble && definition 2008-05-28
Ian Lynagh [Sat, 12 Apr 2008 10:08:06 +0000 (10:08 +0000)] 
Wibble && definition

10 years agoWe can now use Ordering as it's in ghc-prim
Ian Lynagh [Sat, 12 Apr 2008 21:58:13 +0000 (21:58 +0000)] 
We can now use Ordering as it's in ghc-prim

10 years agoDefine integerToWord64, word64ToInteger, integerToInt64, int64ToInteger
Ian Lynagh [Tue, 25 Mar 2008 20:28:04 +0000 (20:28 +0000)] 
Define integerToWord64, word64ToInteger, integerToInt64, int64ToInteger
if on a 32-bit machine

10 years agoFix parsing precedence problem
Ian Lynagh [Sun, 23 Mar 2008 20:29:27 +0000 (20:29 +0000)] 
Fix parsing precedence problem

10 years agoInitial commit of integer(-gmp) package; code copied from base
Ian Lynagh [Sun, 23 Mar 2008 18:34:34 +0000 (18:34 +0000)] 
Initial commit of integer(-gmp) package; code copied from base