[nightly] 03-Apr-2009 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 Fri Apr 3 18:00:01 BST 2009. checking out

[nightly] 03-Apr-2009 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com)

Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location= /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Fri Apr 3 19:00:01 BST 2009. ***

patch applied (testsuite): Add a compiler space usage test for trac #1969

Fri Apr 3 18:09:54 PDT 2009 Ian Lynagh * Add a compiler space usage test for trac #1969 A ./tests/ghc-regress/space_leaks/T1969.hs M ./tests/ghc-regress/space_leaks/all.T +34 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090404010954-3fd76-e06b1b61fbc20ba237b

patch applied (testsuite): Add support for compiler -t stats

Fri Apr 3 17:27:30 PDT 2009 Ian Lynagh * Add support for compiler -t stats M ./driver/testglobals.py -1 +2 M ./driver/testlib.py -37 +63 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090404002730-3fd76-5297cc7cf9ba1836f09230a62893711f5401b4ce.gz

patch applied (ghc): Don't use thread local storage on x86/not-Linux

Fri Apr 3 17:34:45 PDT 2009 Ian Lynagh * 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 M ./rts/sm/GCThread.h

patch applied (testsuite): Expand the space_leak expected results to work on more platforms

Fri Apr 3 13:59:33 PDT 2009 Ian Lynagh * Expand the space_leak expected results to work on more platforms M ./tests/ghc-regress/space_leaks/all.T -7 +20 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403205933-3fd76-d1f6d54af28408bbcc3522627996140eca2b23f9.gz

patch applied (testsuite): Define if_not_platform

Fri Apr 3 13:56:44 PDT 2009 Ian Lynagh * Define if_not_platform M ./driver/testlib.py +6 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403205644-3fd76-082c800931e3f36da8919b202bbed28fcb75d627.gz ___ Cvs-ghc mailing

patch applied (testsuite): Define if_wordsize in the driver

Fri Apr 3 11:20:28 PDT 2009 Ian Lynagh * Define if_wordsize in the driver M ./driver/testlib.py +6 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403182028-3fd76-d94cb20584c9141e26240a59c4d99460aefb0abb.gz ___ Cvs-gh

patch applied (testsuite): Improve the testsuite driver support for -t stats, and enhance space_leak_001

Fri Apr 3 11:08:13 PDT 2009 Ian Lynagh * Improve the testsuite driver support for -t stats, and enhance space_leak_001 M ./driver/testglobals.py -5 +7 M ./driver/testlib.py -37 +33 M ./tests/ghc-regress/space_leaks/all.T -2 +8 View patch online: http://darcs.haskell.org/testsuit

patch applied (testsuite): Rewrite the result testing logic in simple_run

Fri Apr 3 09:19:12 PDT 2009 Ian Lynagh * Rewrite the result testing logic in simple_run M ./driver/testlib.py -8 +11 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403161912-3fd76-bb5f785a0ee166a23fc3c89b4b352400c5fca7ad.gz

patch applied (testsuite): Re-indent simple_run in the driver

Fri Apr 3 08:45:18 PDT 2009 Ian Lynagh * Re-indent simple_run in the driver M ./driver/testlib.py -49 +49 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403154518-3fd76-d759d518a5ceebfe459c968ba5120f49fb5a1e65.gz ___

patch applied (testsuite): Define and use opts in simple_run in the driver

Fri Apr 3 08:43:08 PDT 2009 Ian Lynagh * Define and use opts in simple_run in the driver M ./driver/testlib.py -6 +8 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403154308-3fd76-6c36b919493f3b47b48e90b7f5133a9386f55e65.gz

patch applied (ghc): Make some showSDoc's use OneLineMode rather than PageMode

Tue Mar 31 11:19:48 PDT 2009 Ian Lynagh * Make some showSDoc's use OneLineMode rather than PageMode M ./compiler/codeGen/CgClosure.lhs -2 +2 M ./compiler/rename/RnExpr.lhs -1 +1 M ./compiler/typecheck/TcGenDeriv.lhs -2 +2 M ./compiler/utils/Outputable.lhs -2 +13 View patch onl

patch applied (ghc): Make the showSDoc definition more explicit

Tue Mar 31 10:41:03 PDT 2009 Ian Lynagh * Make the showSDoc definition more explicit M ./compiler/utils/Outputable.lhs -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090331174103-3fd76-ae927efe2ab996dc5f03272b7592891f325ebd45.gz ___

Re: [Haskell-cafe] Possible floating point bug in GHC?

Okay, I can confirm the bug is fixed. It's insane this bug did not cause any more problems. Every call into every C function that uses floating point could have been affected (OpenGL, BLAS, etc) On Fri, Apr 3, 2009 at 10:47 PM, Peter Verswyvelen wrote: > Ouch, what a waste of time on my side :-(

patch applied (ghc): eliminate warnings

Fri Apr 3 13:34:29 PDT 2009 d...@eecs.tufts.edu * eliminate warnings M ./compiler/cmm/CmmCallConv.hs -1 +1 M ./compiler/cmm/MkZipCfgCmm.hs -1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090403203429-38771-bdc2537e3c047e5d18f3476178c8e6662efd6a2f.gz __

patch applied (ghc): Debugging by Sesame Street:

Fri Apr 3 13:15:04 PDT 2009 d...@eecs.tufts.edu * 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

patch applied (ghc): Buggy optimizations caused function-call return to share the function's entry point

Tue Mar 31 07:46:39 PDT 2009 d...@eecs.tufts.edu * 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 t

patch applied (ghc): Better handling of node parameter in calling conventions

Wed Mar 25 09:38:15 PDT 2009 d...@eecs.tufts.edu * 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. M ./compiler/cmm/CmmCallConv.hs -4

patch applied (ghc): When calling gc, avoid saving node in static closures

Mon Mar 23 13:47:44 PDT 2009 d...@eecs.tufts.edu * When calling gc, avoid saving node in static closures M ./compiler/codeGen/StgCmmBind.hs -2 +2 M ./compiler/codeGen/StgCmmHeap.hs -2 +3 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090323204744-38771-38d26a8cee53039

patch applied (ghc): Code simplifications due to call/return separation; some improvements to how node argument is managed

Mon Mar 23 13:11:40 PDT 2009 d...@eecs.tufts.edu * Code simplifications due to call/return separation; some improvements to how node argument is managed M ./compiler/cmm/CmmCallConv.hs -34 +2 M ./compiler/cmm/CmmProcPointZ.hs -3 +2 M ./compiler/cmm/MkZipCfgCmm.hs -9 +9 M ./comp

patch applied (ghc): Code simplification due to separate call/return conventions

Mon Mar 23 11:22:14 PDT 2009 d...@eecs.tufts.edu * Code simplification due to separate call/return conventions M ./compiler/cmm/CmmCallConv.hs -18 +30 M ./compiler/cmm/MkZipCfgCmm.hs -6 +5 M ./compiler/cmm/ZipCfgCmmRep.hs -6 +6 View patch online: http://darcs.haskell.org/ghc/_darcs

patch applied (ghc): Calls with and without passing node arguments more clearly separated

Mon Mar 23 10:47:00 PDT 2009 d...@eecs.tufts.edu * Calls with and without passing node arguments more clearly separated M ./compiler/cmm/CmmCallConv.hs -2 +13 M ./compiler/cmm/CmmCvt.hs -2 +2 M ./compiler/cmm/MkZipCfgCmm.hs -3 +3 M ./compiler/cmm/ZipCfgCmmRep.hs -9 +12 M ./c

patch applied (ghc): Another small step: call and return conventions specified separately when making calls

Mon Mar 23 10:28:37 PDT 2009 d...@eecs.tufts.edu * Another small step: call and return conventions specified separately when making calls M ./compiler/cmm/CmmCallConv.hs -2 +1 M ./compiler/cmm/CmmCvt.hs -1 +1 M ./compiler/cmm/MkZipCfgCmm.hs -6 +7 M ./compiler/cmm/ZipCfgCmmRep.h

patch applied (ghc): Small step toward call-conv improvement: separate out calls and returns

Mon Mar 23 10:07:06 PDT 2009 d...@eecs.tufts.edu * Small step toward call-conv improvement: separate out calls and returns M ./compiler/cmm/CmmCallConv.hs -6 +8 M ./compiler/cmm/CmmCvt.hs -2 +2 M ./compiler/cmm/MkZipCfgCmm.hs -5 +5 M ./compiler/cmm/ZipCfgCmmRep.hs -8 +14 M .

Re: [Haskell-cafe] Possible floating point bug in GHC?

Ouch, what a waste of time on my side :-( This bugfix is not mentioned in the "notable bug fixes" here Since this is such a severe bug, I would recommend listing it :) Anyway, I have a very small repro test case now. Will c

Re: [Haskell-cafe] Possible floating point bug in GHC?

On Fri, Apr 03, 2009 at 10:10:17PM +0200, Peter Verswyvelen wrote: > I tried both precise and fast, but that did not help. Compiling to SSE2 > fixed it, since that does not use a floating point stack I guess. You didn't say what version of GHC you are using, but it sounds like this might already b

Re: [Haskell-cafe] Possible floating point bug in GHC?

I tried both precise and fast, but that did not help. Compiling to SSE2 fixed it, since that does not use a floating point stack I guess. I'm preparing a repro test case, but it is tricky since removing code tends to change the optimizations and then the bug does not occur. Does anybody know what

Re: [Haskell-cafe] Possible floating point bug in GHC?

Well this situation can indeed not occur on PowerPCs since these CPUs just have floating point registers, not some weird dual stack sometimes / registers sometimes architecture. But in my case the bug is consistent, not from time to time. So I'll try to reduce this to a small reproducible test cas

Re: [Haskell-cafe] Possible floating point bug in GHC?

Interesting. This could be the cause of a weird floating point bug that has been showing up in the ghc testsuite recently, specifically affecting MacOS/Intel (but not MacOS/ppc). http://darcs.haskell.org/testsuite/tests/ghc-regress/lib/Numeric/num009.hs That test compares the result of t

patch applied (ghc): On x86, use thread-local storage instead of stealing a reg for gct

Fri Apr 3 05:18:16 PDT 2009 Simon Marlow * On x86, use thread-local storage instead of stealing a reg for gct Ignore-this: b132a4c2f56a955adb8537442ca8eb0f Benchmarks show that using TLS instead of stealing a register is better by a few percent on x86, due to the lack of registers.

patch applied (ghc): in the non-threaded RTS, use a static gc_thread structure

Fri Apr 3 05:14:43 PDT 2009 Simon Marlow * in the non-threaded RTS, use a static gc_thread structure Ignore-this: 631365db4f9ec7c87f71e0ac851cdfc M ./rts/sm/GC.c -22 +22 M ./rts/sm/GCThread.h -3 +17 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090403121443-12142

patch applied (nofib): Comments only

Fri Apr 3 04:23:33 PDT 2009 simo...@microsoft.com * Comments only Ignore-this: aa157d79e69764d28163c0e2c84660c7 M ./spectral/fish/Main.hs +3 View patch online: http://darcs.haskell.org/nofib/_darcs/patches/20090403112333-1287e-3e018b7e192284d16fe721dcadc14dce92eec8a3.gz ___

patch applied (nofib): Add notes to Simon-nofib-notes

Fri Apr 3 04:23:18 PDT 2009 simo...@microsoft.com * Add notes to Simon-nofib-notes Ignore-this: 9e96d5e87aa1b6a7e3c481ecc50cf5cd M ./Simon-nofib-notes +4 View patch online: http://darcs.haskell.org/nofib/_darcs/patches/20090403112318-1287e-5ac54a156978da32656a31cc8ff412f7dc6c7154.gz __

patch applied (testsuite): Test Trac #3106

Fri Apr 3 04:11:06 PDT 2009 simo...@microsoft.com * Test Trac #3106 Ignore-this: be8186b921b57a547d85fee373d4 A ./tests/ghc-regress/simplCore/should_compile/T3016.hs M ./tests/ghc-regress/simplCore/should_compile/all.T +1 View patch online: http://darcs.haskell.org/testsuite/_da

patch applied (ghc): small GC optimisation

Fri Apr 3 01:37:08 PDT 2009 Simon Marlow * small GC optimisation M ./rts/sm/Evac.c -2 +2 M ./rts/sm/GCUtils.c -2 +11 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090403083708-12142-7bb210073f9a4693b9c6ce33d95a0a1080241051.gz __

patch applied (ghc): add -funroll-loops when compiling the parallel version of Evac.c too

Thu Apr 2 03:56:43 PDT 2009 Simon Marlow * add -funroll-loops when compiling the parallel version of Evac.c too Ignore-this: c120d5f5ea961e10acad457fd871ba53 M ./rts/Makefile +2 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090402105643-12142-8cabaa16b3aa139e012eaec2

patch applied (ghc): PACKAGE_CPP_OPTS += -DPAPI_INCLUDE_DIR=""

Thu Apr 2 03:55:40 PDT 2009 Simon Marlow * PACKAGE_CPP_OPTS += -DPAPI_INCLUDE_DIR="" Ignore-this: 4136707371784c791c435ae4f562a500 unbreaks ghc-pkg check in an installed GHC M ./rts/Makefile +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090402105540-12142-85d1ff

patch applied (testsuite): Eyeball test for Trac #3116

Fri Apr 3 03:48:39 PDT 2009 simo...@microsoft.com * Eyeball test for Trac #3116 Ignore-this: e7791ffbc7e31794d2e53c47e85c0e3b A ./tests/ghc-regress/eyeball/T3116.hs View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20090403104839-1287e-53455c05aea1267f5a385654a3ed7127

patch applied (ghc): Adjust inlining heursitics

Fri Apr 3 01:46:34 PDT 2009 simo...@microsoft.com * Adjust inlining heursitics Ignore-this: a2d6849726f6705e06941c74137bfe7e 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 wa

patch applied (ghc): Worker/wrapper should make INLINE if it doesn't w/w

Fri Apr 3 01:43:33 PDT 2009 simo...@microsoft.com * Worker/wrapper should make INLINE if it doesn't w/w Ignore-this: 5ef3c02c272ad59616fa7dfaa0dc0381 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 i

patch applied (ghc): Rewrite a good chunk of CoreArity

Fri Apr 3 01:36:55 PDT 2009 simo...@microsoft.com * Rewrite a good chunk of CoreArity Ignore-this: 314d26ad8f2d4437e11148b58e35d731 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

patch applied (ghc): Comments only; record remarks about removing 'type' nonterminal

Thu Apr 2 09:52:26 PDT 2009 simo...@microsoft.com * Comments only; record remarks about removing 'type' nonterminal Ignore-this: 82f93bd161ee48c0a745fb58247afec0 M ./compiler/parser/Parser.y.pp -11 +37 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20090402165226-1287e-e

patch applied (ghc): Don't float case expressions in full laziness

Thu Apr 2 09:51:01 PDT 2009 simo...@microsoft.com * Don't float case expressions in full laziness Ignore-this: 356f0fc77301283116a669ed76478c4 See Note [Case MFEs]; don't float case expressions from a strict context. M ./compiler/simplCore/SetLevels.lhs +12 View patch online:

patch applied (ghc): Fix Trac #3118: missing alternative

Thu Apr 2 08:28:34 PDT 2009 simo...@microsoft.com * Fix Trac #3118: missing alternative Ignore-this: 9a9866ad88eae871cb98c6795c12349b This patch fixes a rather obscure bug, whereby it's possible for (case C a b of ) to have altenatives that do not inclue (C a b)! See Note [Unreachab

patch applied (ghc): Use a local interestingDict function instead of importing SimplUtils.interestingArg

Thu Apr 2 08:22:46 PDT 2009 simo...@microsoft.com * Use a local interestingDict function instead of importing SimplUtils.interestingArg Ignore-this: d01d1a42067c2f907a1f2fb2ddec4ada I'm changing the details of SimplUtils.interstingArg, and don't want to mess up the way Specialise wor

patch applied (ghc): Better panic reporting

Thu Apr 2 08:18:46 PDT 2009 simo...@microsoft.com * Better panic reporting Ignore-this: abbd0ab8e2e8771012c17e0fc66fed41 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. M ./comp

patch applied (ghc): Fix pragmas in binder-swapping

Thu Apr 2 08:16:16 PDT 2009 simo...@microsoft.com * Fix pragmas in binder-swapping Ignore-this: 422672152247c7732c56e850f3317546 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 IN

patch applied (ghc): Float constants to top-level even in first full laziness pass

Thu Apr 2 08:14:06 PDT 2009 simo...@microsoft.com * Float constants to top-level even in first full laziness pass Ignore-this: c81a6d62b760d50348aaa5de0276d7c6 This patch changes the compilation pipeline flags so that constants are floated to the top level even in the first full lazine

patch applied (testsuite): Test Trac #3118

Thu Apr 2 23:30:30 PDT 2009 simo...@microsoft.com * Test Trac #3118 Ignore-this: a3c9ed32ca5b2438b9eafd96d1d8a691 A ./tests/ghc-regress/simplCore/should_compile/T3118.hs M ./tests/ghc-regress/simplCore/should_compile/all.T +1 View patch online: http://darcs.haskell.org/testsuite/_da

Daily report for stable

Build results: kili stable: pass sparky stable: pass tnaur PPC OSX stable 2: pass tnaur x86 Linux stable: pass x86 Linux stable:lost x86 Windows stable: pass lost x86 Windows stable fast: pass pass pass pass pass lost x86-64 Linux stable: pass Old unexpect

[nightly] 02-Apr-2009 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 Thu Apr 2 18:00:01 BST 2009. checking out