[nightly] 10-Aug-2007 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/ghc/nightly/HEAD-cam-02-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Fri Aug 10 19:30:00 BST 2007. checki

patch applied (testsuite): Update standalone deriving test, and add a failure test

Fri Aug 10 15:34:39 PDT 2007 Ian Lynagh <[EMAIL PROTECTED]> * Update standalone deriving test, and add a failure test M ./tests/ghc-regress/deriving/should_compile/drv014.hs -5 +5 M ./tests/ghc-regress/deriving/should_fail/all.T +1 A ./tests/ghc-regress/deriving/should_fail/drvfail0

patch applied (ghc): Remove a duplicate case

Fri Aug 10 04:57:44 PDT 2007 Ian Lynagh <[EMAIL PROTECTED]> * Remove a duplicate case M ./compiler/ghci/ByteCodeAsm.lhs -1 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc

patch applied (ghc): Change standalone deriving syntax and semantics; fixes trac #1481

Fri Aug 10 15:37:28 PDT 2007 Ian Lynagh <[EMAIL PROTECTED]> * Change standalone deriving syntax and semantics; fixes trac #1481 You now say deriving instance Cxt => Head M ./compiler/parser/Lexer.x -3 M ./compiler/parser/Parser.y.pp -3 +1 M ./compiler/typecheck/TcDeriv.lhs -

patch applied (testsuite): Accept changes (mostly paren removal)

Fri Aug 10 12:44:02 PDT 2007 Ian Lynagh <[EMAIL PROTECTED]> * Accept changes (mostly paren removal) M ./tests/ghc-regress/gadt/gadt13.stderr -1 +1 M ./tests/ghc-regress/parser/should_fail/read023.stderr -1 +1 M ./tests/ghc-regress/rename/should_compile/rn049.stderr -1 +1 M ./tes

Re: Fixed OSX head breakage; was: Build RTS as dynamic library - breaks HEAD on OSX

At Fri, 10 Aug 2007 18:54:37 +0200, Thorkil Naur <[EMAIL PROTECTED]> wrote: > Continuing our conversation on #ghc: I verified after all, using a copy of > your patch that started like the attached file > therp-repair-patch-copy-2.dpatch. Thanks for your time! > Comparing to your version of th

Re: patch applied (ghc): Build RTS as dynamic library - breaks HEAD on OSX

Hello, On Friday 10 August 2007 14:19, Clemens Fruhwirth wrote: > At Fri, 10 Aug 2007 01:25:06 -0700, > Simon Marlow wrote: > > > > Tue Aug 7 23:50:43 PDT 2007 Clemens Fruhwirth <[EMAIL PROTECTED]> > > * Build RTS as dynamic library > > > > M ./rts/Makefile -38 +35 > > Switching to the N

patch applied (ghc): Properly guard imports because they have to be precise on Windows and Darwin sets __PIC__ automatically

Fri Aug 10 08:10:45 PDT 2007 Clemens Fruhwirth <[EMAIL PROTECTED]> * Properly guard imports because they have to be precise on Windows and Darwin sets __PIC__ automatically M ./rts/Exception.cmm +2 M ./rts/HeapStackCheck.cmm +2 M ./rts/PrimOps.cmm +2 M ./rts/StgMiscClosures.cmm

patch applied (ghc): Add explicit imports for RTS-external variables

Mon Aug 6 01:49:38 PDT 2007 Clemens Fruhwirth <[EMAIL PROTECTED]> * Add explicit imports for RTS-external variables M ./rts/Exception.cmm +2 M ./rts/HeapStackCheck.cmm +2 M ./rts/PrimOps.cmm +20 M ./rts/StgMiscClosures.cmm -4 +7 ___

Broken HEAD

To be precise, this is what I'm getting from validate on x86 linux with current HEAD, in case it matters. I think if no one offers an opinion to the contrary, I'll feel okay to commit if validate isn't broken any worse than HEAD, unless my patches are in the same area as some of the breakage o

Re: patch applied (ghc): Build RTS as dynamic library - breaks HEAD on OSX

At Fri, 10 Aug 2007 01:25:06 -0700, Simon Marlow wrote: > > Tue Aug 7 23:50:43 PDT 2007 Clemens Fruhwirth <[EMAIL PROTECTED]> > * Build RTS as dynamic library > > M ./rts/Makefile -38 +35 Switching to the NCG for the RTS build breaks HEAD on OSX. The cmm source does not declare which symb

RE: My current speed-testing validating technique

I can't comment about the build process (others will) but if you find performance regressions please do report them as precisely as possible when you have them nailed down, so we can reproduce them. Thanks S | -Original Message- | From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] | On

My current speed-testing validating technique

The procedure I am about to describe, should make me notice serious performance regressions in the compiled stage1. Also it tells me when my patches don't break validate any *more* than an unpatched HEAD (***which is true currently, that validate is slightly broken for me. Should I commit? S

hppa/HPUX registerised AND optimized port of GHC 6.6.1 available

I was able to fix the gcc optimizer and have rebuilt ghc 6.6.1 for hpux-hppa with the optimizer enabled. The compiler binary is half the size of the unoptimized binary. I have placed binaries, build scripts, build logs, patches etc. at: http://72.196.247.179:3000/ghc/optimized/ Joe Buehler ___

patch applied (ghc): Improve error message (part of Trac #1606)

Fri Aug 10 01:33:12 PDT 2007 [EMAIL PROTECTED] * Improve error message (part of Trac #1606) M ./compiler/typecheck/TcMatches.lhs -1 +8 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc

patch applied (testsuite): Test Trac #1606

Fri Aug 10 02:03:46 PDT 2007 [EMAIL PROTECTED] * Test Trac #1606 M ./tests/ghc-regress/typecheck/should_fail/all.T +1 A ./tests/ghc-regress/typecheck/should_fail/tcfail185.hs A ./tests/ghc-regress/typecheck/should_fail/tcfail185.stderr __

patch applied (ghc): Print fewer parens when pretty-printing HsSyn

Thu Aug 9 15:33:29 PDT 2007 [EMAIL PROTECTED] * Print fewer parens when pretty-printing HsSyn M ./compiler/hsSyn/HsExpr.lhs -6 +21 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc

patch applied (ghc): Print more "..." in long lists of declarations

Thu Aug 9 15:32:28 PDT 2007 [EMAIL PROTECTED] * Print more "..." in long lists of declarations M ./compiler/hsSyn/HsBinds.lhs -3 +3 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/mailman/listinfo/cvs-ghc

patch applied (ghc): Build RTS as dynamic library

Tue Aug 7 23:50:43 PDT 2007 Clemens Fruhwirth <[EMAIL PROTECTED]> * Build RTS as dynamic library M ./configure.ac +7 M ./includes/StgDLL.h -13 +13 M ./mk/config.mk.in +18 M ./mk/package.mk -63 +4 M ./mk/target.mk -61 +38 M ./rts/Linker.c -6 +2 M ./rts/Makefile -38 +

patch applied (ghc): Introduce new class for external symbols in Linker.c that use __imp__ instead of &

Mon Aug 6 01:44:49 PDT 2007 Clemens Fruhwirth <[EMAIL PROTECTED]> * Introduce new class for external symbols in Linker.c that use __imp__ instead of & M ./rts/Linker.c -6 +18 ___ Cvs-ghc mailing list Cvs-ghc@haskell.org http://www.haskell.org/m

Re: patch applied (ghc): Implicitly import NDP stuff when vectorising

Simon Peyton-Jones wrote: Interesting. Are you sure you *want* to refer to NDP stuff by RdrName? Polluting the GblRdrEnv sounds suspicious to me That was your suggestion! :-) Remember the problem of getting at built-in PA dictionaries for certain types? Originally, we wanted to use a class

RE: patch applied (ghc): Implicitly import NDP stuff when vectorising

Interesting. Are you sure you *want* to refer to NDP stuff by RdrName? Polluting the GblRdrEnv sounds suspicious to me Simon | -Original Message- | From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Roman Leshchinskiy | Sent: 10 August 2007 04:13 | To: cvs-ghc@haskell.org