- ghc-devs - Haskell.org

perf.haskell.org’s list of branches got more useful
by Joachim Breitner 27 Apr '16

27 Apr '16

Hi, the list of branches on https://perf.haskell.org/ghc got more useful. It now shows the difference from the branch head to the merge base¹, which should roughly exhibit the performance change that would be caused by merging the branch into master. If the branch head or the merge base have not been measured (e.g. because the build failed), the display might be a bit confusing. Same with branches that often merge or cherry-pick from master. But in the common case of a linear, possibly rebased, feature branch, it should work fine. Greetings, Joachim ¹ Which, I think, is the youngest commit also reachable from master. Or something like that. Consult "man git merge-base" if you want a little mind teaser. -- Joachim “nomeata” Breitner mail(a)joachim-breitner.de • https://www.joachim-breitner.de/ XMPP: nomeata(a)joachim-breitner.de • OpenPGP-Key: 0xF0FBF51F Debian Developer: nomeata(a)debian.org

4 9

Can we do something slightly nicer about (^)?
by David Feuer 26 Apr '16

26 Apr '16

Every time someone writes, say, x^20, the literal exponent defaults to Integer. This is the wrong default whenever the literal is in the Word range. Fixing this goes beyond the capabilities of RULES pragmas, but I imagine it would be a fairly simple thing to accomplish in the internal rule language. David

3 2

gcc5 and apple clang variants of mac ghc 8.0 rc4 build
by Carter Schonwald 25 Apr '16

25 Apr '16

since i've yet to dig into if any libgcc symbols ever get invoked, heres versions of the RC for mac built using gcc5 and apple clang respectively (in addition to the other one i've already provided that has a gcc6 libgcc dylib that seems to spuriously leak through) https://wellposed-carter-files.s3.amazonaws.com/opensource/ghc/releasebuild… shasum -a512 vanilla-clang-ghc-8.0.0.20160421-x86_64-apple-darwin.tar.xz 9105b097bb8036d6a77fc7bb87bcb0efff4ac7136232ac7ce109664ecae6e0e6757904f4a80de9120813530b979e7d90a36b4b28a1ef7bb9c371ab10a346c9dc vanilla-clang-ghc-8.0.0.20160421-x86_64-apple-darwin.tar.xz http://wellposed-carter-files.s3.amazonaws.com/opensource/ghc/releasebuild-… shasum -a512 vanilla-gcc5-ghc-8.0.0.20160421-x86_64-apple-darwin.tar.xz 43bb84fe73ac9e450d73a877022ba2ca514a08f69c18cac7367ed7038e9f7371b21cdaeef635aee345f9ba001a2fb5b527975e790f1b8d1c6f91bdf231e05090 vanilla-gcc5-ghc-8.0.0.20160421-x86_64-apple-darwin.tar.xz enjoy! -Carter

1 0

reproducible crash in profiled builds when -hb or -hr are enabled in ghc 8.0rc4 (and 7.10.3 and older)
by Carter Schonwald 24 Apr '16

24 Apr '16

Hello all, I've found a reproducible (on OS X) crash when using biographical or retainer profiling (different crashes respectively) that doesn't happen otherwise on profiled builds executables, https://ghc.haskell.org/trac/ghc/ticket/11978 should provide enough information to replicate it (though by no means minimal :(, but if need be i'll try to help out on that front) i've marked it as "highest" because an otherwise ok program triggers a sigsegv with -hb! (which may or may not be related to the different crash that -hr triggered). I'm not sure how to best go about debugging this sort of crash (though if i can find time later this week i'm happy to help) many thanks -Carter

1 0

Re: GHC 8.0.1-rc3 source tarball availability
by Carter Schonwald 24 Apr '16

24 Apr '16

Downloading the ttf font files and opening them via font book handles the import / font registry on Macs On Apr 13, 2016 3:56 PM, "Austin Seipp" <austin(a)well-typed.com> wrote: If you look at the bottom of the error, `fontspec` has already given you the more specific diagnosis of what's wrong: > The font "DejaVu Sans" cannot be found. That is, you need to install the DejaVu Sans font. Dealing with fonts and modern typesetting is a complete mystery to me though, so I'll leave you to install it on your Mac. :) On Wed, Apr 13, 2016 at 2:49 PM, George Colpitts <george.colpitts(a)gmail.com> wrote: > I got past my first problem (see > https://ghc.haskell.org/trac/ghc/ticket/11823#comment:13 for the resolution) > but now I have the following problem building the doc (on a Mac) . > > It seems like it doesn't like font names with spaces, not sure why (or if!) > this would be Mac specific: > > Invalid fontname `DejaVu Sans', contains ' ' > > longer extract from the output: > > copying static files... This is XeTeX, Version 3.14159265-2.6-0.99992 (TeX > Live 2015) (preloaded format=xelatex) > restricted \write18 enabled. > entering extended mode > (./users_guide.tex > LaTeX2e <2015/01/01> > Babel <3.9l> and hyphenation patterns for 79 languages loaded. > (./sphinxmanual.cls > Document Class: sphinxmanual 2009/06/02 Document class (Sphinx manual) > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/report.cls > Document Class: report 2014/09/29 v1.4h Standard LaTeX document class > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/size10.clodone > copying extra files... done > dumping search index in English (code: en) ... ))) > (/usr/local/texlive/2015/texmf-dist/tex/latex/cmap/cmap.sty > > Package cmap Warning: pdftex not detected - exiting. > > )done > dumping object inventory... > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/fontenc.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/t1enc.defdone > build succeeded, 25 warnings. > )) > (/usr/local/texlive/2015/texmf-dist/tex/generic/babel/babel.sty > (/usr/local/texlive/2015/texmf-dist/tex/generic/babel-english/english.ldf > (/usr/local/texlive/2015/texmf-dist/tex/generic/babel/babel.def > (/usr/local/texlive/2015/texmf-dist/tex/generic/babel/xebabel.def)))) > (/usr/local/texlive/2015/texmf-dist/tex/latex/psnfss/times.sty) > (./fncychap.sty > ) (/usr/local/texlive/2015/texmf-dist/tex/latex/tools/longtable.sty) > (./sphinx.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/fancyhdr/fancyhdr.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/textcomp.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/ts1enc.def)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/fancybox/fancybox.sty > Style option: `fancybox' v1.4 <2010/05/15> (tvz) > ) (/usr/local/texlive/2015/texmf-dist/tex/latex/titlesec/titlesec.sty) > (./tabulary.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/tools/array.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/amsmath/amsmath.sty > For additional information on amsmath, use the `?' option. > (/usr/local/texlive/2015/texmf-dist/tex/latex/amsmath/amstext.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/amsmath/amsgen.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/amsmath/amsbsy.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/amsmath/amsopn.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/makeidx.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/framed/framed.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/ifthen.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/graphics/color.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/latexconfig/color.cfg) > (/usr/local/texlive/2015/texmf-dist/tex/xelatex/xetex-def/xetex.def)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/fancyvrb/fancyvrb.sty > Style option: `fancyvrb' v2.7a, with DG/SPQR fixes, and firstline=lastline > fix > <2008/02/07> (tvz) > (/usr/local/texlive/2015/texmf-dist/tex/latex/graphics/keyval.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/threeparttable/threeparttable.sty > ) (/usr/local/texlive/2015/texmf-dist/tex/latex/mdwtools/footnote.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/wrapfig/wrapfig.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/parskip/parskip.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/alltt.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/upquote/upquote.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/graphics/graphicx.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/graphics/graphics.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/graphics/trig.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/latexconfig/graphics.cfg))) > (/usr/local/texlive/2015/texmf-dist/tex/latex/hyperref/hyperref.sty > (/usr/local/texlive/2015/texmf-dist/tex/generic/oberdiek/hobsub-hyperref.sty > (/usr/local/texlive/2015/texmf-dist/tex/generic/oberdiek/hobsub-generic.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/generic/ifxetex/ifxetex.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/oberdiek/auxhook.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/oberdiek/kvoptions.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/hyperref/pd1enc.def) > (/usr/local/texlive/2015/texmf-dist/tex/latex/latexconfig/hyperref.cfg) > (/usr/local/texlive/2015/texmf-dist/tex/latex/url/url.sty)) > > Package hyperref Message: Driver (autodetected): hxetex. > > (/usr/local/texlive/2015/texmf-dist/tex/latex/hyperref/hxetex.def > (/usr/local/texlive/2015/texmf-dist/tex/latex/hyperref/puenc.def) > (/usr/local/texlive/2015/texmf-dist/tex/generic/oberdiek/stringenc.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/oberdiek/rerunfilecheck.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/oberdiek/hypcap.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/caption/newfloat.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/capt-of/capt-of.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/needspace/needspace.sty)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/multirow/multirow.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/eqparbox/eqparbox.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/environ/environ.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/trimspaces/trimspaces.sty))) > (/usr/local/texlive/2015/texmf-dist/tex/latex/fontspec/fontspec.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/l3kernel/expl3.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/l3kernel/expl3-code.tex) > (/usr/local/texlive/2015/texmf-dist/tex/latex/l3kernel/l3unicode-data.def) > (/usr/local/texlive/2015/texmf-dist/tex/latex/l3kernel/l3xdvipdfmx.def)) > (/usr/local/texlive/2015/texmf-dist/tex/latex/l3packages/xparse/xparse.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/fontspec/fontspec-patches.sty) > (/usr/local/texlive/2015/texmf-dist/tex/latex/fontspec/fontspec-xetex.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/base/fontenc.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/euenc/eu1enc.def) > (/usr/local/texlive/2015/texmf-dist/tex/latex/euenc/eu1lmr.fd)) > (/usr/local/texlive/2015/texmf-dist/tex/xelatex/xunicode/xunicode.sty > (/usr/local/texlive/2015/texmf-dist/tex/latex/tipa/t3enc.def > (/usr/local/texlive/2015/texmf-dist/tex/latex/euenc/eu1lmss.fd))) > (/usr/local/texlive/2015/texmf-dist/tex/latex/fontspec/fontspec.cfg)))kpathsea:make_tex: > . > > > !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! > ! > ! fontspec error: "font-not-found" > ! > ! The font "DejaVu Sans" cannot be found. > ! > ! See the fontspec documentation for further information. > ! > ! For immediate help type H <return>. > !............................................... > > l.23 \setsansfont{DejaVu Sans} > > ? > > On Sun, Apr 10, 2016 at 12:05 PM Ben Gamari <ben(a)well-typed.com> wrote: >> >> tl;dr: If you would like to produce a binary distribution for GHC >> 8.0.1-rc3 then let us know, grab the source distribution and >> start building. The binary distributions will be all released one >> week from today. >> >> >> Hello GHC packagers, >> >> Austin and I are happy to announce the release of the 8.0.1-rc3 source >> distribution to binary packagers. You will find the tarball and usual >> assortment of signatures at >> >> http://downloads.haskell.org/~ghc/8.0.1-rc3/ >> >> This release is the first where we are following our new release policy >> [1], where a one-week delay will be introduced between the release of >> the source and binary distributions. The goal of this policy is to give >> all platforms the opportunity for support from the first day of a >> release. >> >> If all of the expected binary releases are submitted before the >> week-long delay is through, we can move forward with the release ahead >> of schedule. Given how late release candidate 3 is, we'd naturally like >> to get the binary releases out as soon is possible: The sooner we can >> get this -rc out the door the sooner we can get to the final release. >> Moreover, it would be appreciated if you could confirm that you intend >> to offer a binary distribution this release. >> >> Otherwise, let either Austin or I know if you have any trouble building >> your distribution. I have yet to push the 8.0.1-rc3 tag in case we >> encounter unexpected issues but all of my builds with this tarball >> thusfar have gone quite well. >> >> Good luck and thanks for all of your work! >> >> Cheers, >> >> - Ben >> >> >> [1] https://mail.haskell.org/pipermail/ghc-devs/2016-March/011546.html >> _______________________________________________ >> ghc-devs mailing list >> ghc-devs(a)haskell.org >> http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs > > > _______________________________________________ > ghc-devs mailing list > ghc-devs(a)haskell.org > http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs > -- Regards, Austin Seipp, Haskell Consultant Well-Typed LLP, http://www.well-typed.com/ _______________________________________________ ghc-devs mailing list ghc-devs(a)haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

4 4

make target doesn't include make pdf or html targets by default!
by Carter Schonwald 23 Apr '16

23 Apr '16

even though i had sphinx html and pdf flags set to YES in my mk/build.mk file! this was quite surprising. This seems to have been a change that happened sometime after the RC3 tag...

2 1

[ANNOUNCE] GHC 8.0.1 release candidate 3 available
by Ben Gamari 23 Apr '16

23 Apr '16

Hello Haskellers, The GHC team is happy to announce the availability of third release candidate of the Glasgow Haskell Compiler's 8.0.1 release. Source and binary distributions can be found at, http://downloads.haskell.org/~ghc/8.0.1-rc3/ This is the third of what will be four candidates leading up to the 8.0.1 release. As there are a number of known issues with this candidate, we expect to produce the fourth and final release candidate shortly. While we considered not releasing this candidate at all and instead skipping to rc4, we decided that it would be helpful to make it available despite its issues to increase the likelihood of catching any remaining bugs. That being said, this candidate lacks the range of binary distributions which we usually offer, * We are unable to release Windows bindists due to a compatibility issue which we didn't catch until after preparing the source distribution. * Binaries targeting Redhat distributions are currently not available Regardless, users for whom a distribution is available are encouraged to try this candidate as it fixes dozens of issues present in release candidate two. These include, * A type-checker panic triggered by use of Typeable on a kind-polymorphic type constructor (#11334) * A type-checker explosion where -XTypeInType would gobble up massive amounts of memory when used in a data instance (#11407) * Various issues involving unexpected laziness resulting in exception handlers not being invoked (#11555) * GHC now fails more gracefully when used with an older cabal-install release (#11558) * TypeInType now has proper documentation in the users guide (#11614) * The story surrounding type `RuntimeRep`s (formerly known as `Levity`) is now far better developed, closing several doors to unsafe behavior that TypeInType previously opened (#11473, #11723) * A long-standing bug in the constant-folding rules for `mod` for the `Word` type has been resolved (#11702) * Various issues introduced by OverloadedRecordFields have been fixed (#11662, #11401) * A regression in the typechecker resulting in the rejection of code in the `free` and `microlens` packages has been fixed (#11608) * A bug in the LLVM code generator which caused some programs to emit unbuildable LLVM IR has been fixed (#11649) * A bug where pattern synonyms defined in terms of a pattern match on a record would be rejected if the fields weren't written in the same order as they were defined has been resolved (#11633) * A bug in the runtime system's treatment of weak references which could result in segmentation faults in some cases has been fixed (#11108) * a variety of optimizations improving compiler performance have been merged * and many others Mac OS X users should be aware that the recent XCode 7.3 release ships with a broken `nm` utility which breaks the GHC build (#11744, #11823). Unfortunately, this candidate does not notice this brokenness and odd behavior can result. OS X users running XCode 7.3 are encouraged to open a ticket with Apple and build GHC with the `--with-nm=nm-classic` configure option until a fix is available. As mentioned earlier, there are a number of known issues with this candidate which will be addressed in release candidate four. These issues include, * A number of typechecker bugs (#11811, #11797, #11813, #11814) * A build issue seen on OS X (#11828) * Template Haskell lacks the ability to produce instances with OVERLAPP{ING,ABLE,ED} pragmas * Autoconf has inadequate version checks for libdw (#11820) * Typeable and hs-boot files interact poorly (#11824) * The build system doesn't check for the broken Apple `nm` utility, leading to unpredictable results (#11823, #11744) We hope to have release candidate four available by the middle of next week. Thanks to everyone who has contributed code, testing, builds, and bug reports thusfar! Cheers, - Ben

2 1

Accessing dictionaries from a GHC plugin
by Yannick Boesmans 22 Apr '16

22 Apr '16

Hi, I'm trying to develop a GHC plugin to transform effect handlers for free monads. I have created my own module with the free monad definitions. The goal is that if a user uses these definitions, my plugin will rewrite the handlers into a form which will allow fusion. For a first simple example I'm trying to transform the following code: handleNondet :: Functor g => Free (Nondet + g) a -> Free g [a] handleNondet = \y -> case y of Var x -> Var [x] Con op -> case op of Inl x -> case x of Or l r -> (handleNondet l >>= (\ll -> handleNondet r >>= (\rr -> Var (ll ++ rr)))) Inr x -> Con (fmap handleNondet x) into this: handleNondet :: Functor g => Free (Nondet + g) a -> Free g [a] handleNondet = foldFree genNondet algNondet genNondet :: Functor g => a -> Free g [a] genNondet x = Var [x] algNondet :: Functor g => (Nondet + g) (Free g [a]) -> Free g [a] algNondet op = case op of Inl x -> case x of Or l r -> (l >>= (\ll -> r >>= (\rr -> Var (ll ++ rr)))) Inr x -> Con x ----------------------------------------------------------------------------------------------------------------------------------- In my module with definitions, Free, Con, Var, the (+), Inl and Inr are defined and the foldFree function is defined. There is also a functor instance: instance (Functor f, Functor g) => Functor (f + g) The problem is that when trying to do this transformation in GHC Core, I only get a dictionary for the functor g, but not for Nondet or for (Nondet + g). What is need is of the form: (WhatMorphism.Free.$fFunctor+ @ Nondet @ g_a2MS Types.$fFunctorNondet $dFunctor_a2R9) and I have access to @ Nondet, @ g_a2MS and $dFunctor_a2R9, but I don't know how to find WhatMorphism.Free.$fFunctor+ and Types.$fFunctorNondet. I am trying to find out whether it is possible to find these dictionaries from a GHC Core plugin, or whether this has to happen in an earlier stage of the compilation. If it is possible, how would I go about it? And if it is not, where should I start looking to implement this functionality anyway? Regards, Yannick Boesmans

2 1

CmmSwitchTest
by Alex Dzyoba 22 Apr '16

22 Apr '16

Hello, I was trying to reproduce issue #11297 (CmmSwitchTest is broken on 32-bit platforms ) from Newcomers page and seems to stuck. I would like to ask for help here, but if there is a more suitable place to ask, please let me know. I've built current master (8c33cd4) stage2 compiler, removed "when(wordsize(32), expect_broken(11297))" from testsuite/tests/codeGen/should_run/all.T to see same errors as in issue and got this (unrelated?) assertion: =====> CmmSwitchTest(normal) 1 of 1 [0, 0, 0] cd ./codeGen/should_run && "/home/avd/ghc/inplace/test spaces/ghc-stage2" -o CmmSwitchTest CmmSwitchTest.hs -fforce-recomp -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-warn-tabs -fno-warn-missed-specialisations -fshow-warning-groups -fno-ghci-history > CmmSwitchTest.comp.stderr 2>&1 Compile failed (status 256) errors were: [1 of 1] Compiling Main ( CmmSwitchTest.hs, CmmSwitchTest.o ) ghc-stage2: panic! (the 'impossible' happened) (GHC version 8.1.20160419 for i386-unknown-linux): ASSERT failed! CallStack (from HasCallStack): assertPprPanic, called at compiler/basicTypes/Literal.hs:224:75 in ghc:Literal 18446744073709551615 Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug *** unexpected failure for CmmSwitchTest(normal) Is there anything I can do with this? Is it because that test has 64-bit unboxed integers and I'm running it on 32-bit VM? Thanks, Alex

2 2

Possible compile time regression in GHC 8
by Sreenidhi Nair 21 Apr '16

21 Apr '16

Hello, While compiling the `record` package with ghc-8 rc3 with `stack build`, the memory usage went above 6gb and started trashing. But when `stack build --fast` (-O0) was being used the package compiled with maximum memory at 1.8gb. Also `stack build` works fine with ghc-7.10.3 with the build taking about 1.7gb. Attached is the stack file to reproduce the issue. Steps to reproduce: 1. Create a new directory and put the attached stack.yaml there. 2. run `stack build` - this will thrash after some time 3. run `stack build --fast` - this will succeed 4. run `stack build --resolver=ghc-7.10.3` - this will succeed as well

2 2