patch applied (ghc): Teach cheapEqExpr about casts

Tue Feb 5 19:50:07 PST 2008 Roman Leshchinskiy <[EMAIL PROTECTED]> * Teach cheapEqExpr about casts Previously, cheapEqExpr would always return False if it encountered a cast. This was bad for two reasons. Firstly, CSE (which uses cheapEqExpr to compare expressions) never eliminated exp

[nightly] 05-Feb-2008 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com)

Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location= /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 5 18:10:01 GMT 2008. checkin

[nightly] 05-Feb-2008 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com)

Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location= /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 5 18:00:01 GMT 2008. checking out

Darcs patch: let runghc.hs read from stdin if no file is provided

This is a rather fun patch (imo) that lets runghc read from stdin (into a temp file) if no filename is provided. It means this works, $ cat A.hs | runghc | python2.4| ruby | runghc | python2.4 | ruby Where A.hs is sigfpe's 3rd order quine, http://sigfpe.blogspot.com/2008/02/third-or

patch applied (ghc): Inject implicit bindings before the simplifier (Trac #2070)

Tue Feb 5 08:55:07 PST 2008 [EMAIL PROTECTED] * Inject implicit bindings before the simplifier (Trac #2070) With constructor unpacking, it's possible for constructors and record selectors to have non-trivial code, which should be optimised before being fed to the code generator. Examp

patch applied (ghc): Make do-notation a bit more flexible (Trac #1537)

Tue Feb 5 08:48:16 PST 2008 [EMAIL PROTECTED] * Make do-notation a bit more flexible (Trac #1537) This is a second attempt to fix #1537: to make the static typechecking of do-notation behave just like the desugared version of the same thing. This should allow parameterised monads to wo

patch applied (ghc): White space only

Tue Feb 5 08:37:02 PST 2008 [EMAIL PROTECTED] * White space only M ./compiler/iface/LoadIface.lhs -1 +2 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc

patch applied (testsuite): add test for #2047

Tue Feb 5 08:52:21 PST 2008 Simon Marlow <[EMAIL PROTECTED]> * add test for #2047 A ./tests/ghc-regress/rts/2047.hs M ./tests/ghc-regress/rts/all.T +3 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-g

patch applied (testsuite): accept output (no newline after uncaught exception)

Tue Feb 5 08:52:12 PST 2008 Simon Marlow <[EMAIL PROTECTED]> * accept output (no newline after uncaught exception) M ./tests/ghc-regress/lib/Process/process004.stderr-i386-unknown-mingw32 -1 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://ww

patch applied (ghc): FIX #2047: Windows (and older Unixes): align info tables to 4 bytes, not 2

Tue Feb 5 02:14:25 PST 2008 Simon Marlow <[EMAIL PROTECTED]> * FIX #2047: Windows (and older Unixes): align info tables to 4 bytes, not 2 Perhaps in the past '.align 2' meant align to 4 bytes, but nowadays it means align to 2 bytes. The compacting collector requires info tables to be ali

RE: Strictness in UniqFM

| I'll take a deeper look at making it strict. I believe that if you do so, you'll _definitely_ need to pay attention to LoadIface.loadDecl. Apart from that, making UniqFM strict would probably be fine. To test this hypothesis you could use a different map for type environments, for now. Wit

Re: Strictness in UniqFM

On Tue, Feb 05, 2008 at 10:51:29AM +, Simon Peyton-Jones wrote: > > hunk ./compiler/utils/UniqFM.lhs 701 > - | j ==# i = mkLeafUFM j (f old new) > + | j ==# i = mkLeafUFM j $! f old new > > This gives UniqFM a very odd behaviour: it becomes strict in the *range* > (element ty

patch applied (ghc): Most of installer for framework on system volume

Mon Feb 4 23:37:38 PST 2008 Manuel M T Chakravarty <[EMAIL PROTECTED]> * Most of installer for framework on system volume M ./distrib/MacOS/GHC-system.pmdoc/01ghc.xml -1 +1 M ./distrib/MacOS/GHC-system.pmdoc/index.xml -1 +3 M ./distrib/MacOS/Makefile +1 A ./distrib/MacOS/instal

patch applied (ghc): Split into two types of Mac installer specs

Mon Feb 4 21:25:04 PST 2008 Manuel M T Chakravarty <[EMAIL PROTECTED]> * Split into two types of Mac installer specs A ./distrib/MacOS/GHC-relocatable.pmdoc/ A ./distrib/MacOS/GHC-relocatable.pmdoc/01ghc-contents.xml A ./distrib/MacOS/GHC-relocatable.pmdoc/01ghc.xml A ./distrib

patch applied (ghc): Lambda logo for packages

Mon Feb 4 21:20:17 PST 2008 Manuel M T Chakravarty <[EMAIL PROTECTED]> * Lambda logo for packages - This image is in the public domain, cf http://en.wikipedia.org/wiki/Image:Greek_lc_lamda_thin.svg A ./distrib/MacOS/lambda-logo.png ___ Cv

patch applied (ghc): xcode build target for fixed /Library/Frameworks inst

Mon Feb 4 19:00:47 PST 2008 Manuel M T Chakravarty <[EMAIL PROTECTED]> * xcode build target for fixed /Library/Frameworks inst - Also moving all MacOS-specific Makefile components into distrib/MacOS/Makefile M ./Makefile -31 +27 M ./compiler/main/SysTools.lhs -1 +7 M ./distr

patch applied (ghc): First stab at an installer package for the Mac

Sat Feb 2 05:48:53 PST 2008 Manuel M T Chakravarty <[EMAIL PROTECTED]> * First stab at an installer package for the Mac - GHC as a Mac framework - I tried to make a package where the user could choose whether to install in /Library/Frameworks or ~/Library/Frameworks (to allow installati

Strictness in UniqFM

Ian, you made this change to UniqFM: Sat Feb 2 21:35:42 GMT Standard Time 2008 Ian Lynagh <[EMAIL PROTECTED]> * Tweak strictness hunk ./compiler/utils/UniqFM.lhs 701 - | j ==# i = mkLeafUFM j (f old new) + | j ==# i = mkLeafUFM j $! f old new This gives UniqFM a very odd be

patch applied (testsuite): accept output (no newline after uncaught exception)

Tue Feb 5 01:48:00 PST 2008 Simon Marlow <[EMAIL PROTECTED]> * accept output (no newline after uncaught exception) M ./tests/ghc-regress/concurrent/prog002/concprog002.stderr -1 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org

Re: Booting from HC files

Matthias Kilian wrote: I'm currently trying to bring back HC bootstrapping, and I'd some kind of success, but only with really horrible hacks on libraries/Makefile.local, the Cabal Makefile template, some bits in mk, and libraries/Makefile itself. FYI the current state is that the HC build start

Re: [nightly] 04-Feb-2008 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com)

GHC Build Reports wrote: Nightly build started on cam-02-unx at Mon Feb 4 18:00:02 GMT 2008. Our x86 compute server is back online after an upgrade, and is now running Fedora 8, which means we'll be able to build some more up-to-date x86/Linux binary dists for the next release. I could ins

patch applied (ghc): Support for using libffi to implement FFI calls in GHCi (#631)

Mon Feb 4 08:10:53 PST 2008 Simon Marlow <[EMAIL PROTECTED]> * Support for using libffi to implement FFI calls in GHCi (#631) This means that an unregisterised build on a platform not directly supported by GHC can now have full FFI support using libffi. Also in this commit: - use

patch applied (ghc): Use the correct libffi type for pointers

Fri Jan 4 05:19:36 PST 2008 Simon Marlow <[EMAIL PROTECTED]> * Use the correct libffi type for pointers M ./compiler/deSugar/DsForeign.lhs -1 +1 M ./rts/Adjustor.c +1 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailma

Daily report for head

Build results: x86-64 Linux head:pass x86 Windows head fast:pass lost pass pass pass gabor head: pass kahl G5 Gentoo Linux head:pass mnemosyne x86-64 Gentoo head: pass tnaur PPC OSX head 2: fail (failed stage2) tnaur x86 Linux head: pass x86

Daily report for stable

Build results: kahl G5 Gentoo Linux stable: pass tnaur PPC OSX stable:pass x86 Windows stable: pass x86 Windows stable fast: pass pass pass pass pass fail (failed stage1) x86-64 Linux stable: pass New unexpected test failures: Chan001 1 x86 Windows stab