- ghc-devs - Haskell.org

GHC 7.8 Release Status & Schedule
by Austin Seipp 10 Oct '13

10 Oct '13

Friends, After talking with Simon yesterday, we have some idea of how the release will go. As I'm sure you're aware, the release is winding down rather quickly, and it will be a fantastic one hopefully :) Now that all the features have landed, we're going into bugfixing mode. The schedule is, roughly: - Nov 1st: Cut branch, and I plan on making a 7.8 RC1 available the same day, for several platforms. - Release will be ~3 weeks out from there, approximately Nov 25th. This gives us a month of solid bugfixing. During the 3 weeks with the 7.8 branch open, I'm a bit hesitant to land massive changes, so that our branches don't diverge too far. OTOH, a few things can probably land in this timeframe with minimal disturbances (such as the Applicative-Monad change.) If you want to land something in that time frame, please just ask me. On the whole, things actually feel pretty good - although there a few nasty bugs to sort out, the uptake in community involvement has simply been fantastic (definitely related to the number bugs we've found,) and I think we're on track to sort the remaining stuff out. A great thanks to all of the new people helping out! And if you want to help even more, please check out the tickets remaining for 7.8.1: http://ghc.haskell.org/trac/ghc/query?status=!closed&milestone=7.8.1&order=… If you think you can take on a bug, please assign it to yourself so we know what's going on. Note: if the bug isn't *high* or *highest*, it's unlikely to get looked at, at least by me! It won't be rejected if you submit a patch, but I'm simply not going to be able to get to it I'm afraid. So please grab something, and go for it! I am also reminded that the GHC October Status report will be due shortly - I'll take some time to write it up, and follow through here on the list (and glasgow-haskell-users) so interested parties can read up on what is happening. -- Austin Seipp, Haskell Consultant Well-Typed LLP, http://www.well-typed.com/

4 3

ghc.haskell.org updated to Trac 1.0.1
by Herbert Valerio Riedel 09 Oct '13

09 Oct '13

Hello *, The Trac instances at ghc.haskell.org have been upgraded to the current stable Trac 1.0.1 release in order to alleviate some technical issues with the aging Trac 0.12.x line. Ideally, you shouldn't notice any regressions; however, as a side-effect Trac 1.0.x has a slightly different look which takes a bit getting used to. I've tried to tweak the navigation sidebar to match the new design and integrate better with Trac's variable-height notification <div/>s. Should you notice anything that has been broken by the upgrade, please let me know! For those interested in what's new in Trac 1.0, here's a subjective selection of Trac 1.0 highlights: - Improved the handling of ticket edit conflicts, by showing simultaneous edits and conflicting changes, and making it easier to merge the changes - Ticket comments can now be displayed threaded - Simplified reverting of a wiki page to an older version - Added a user preference and a option to display relative/absolute datetime - Added //Comments only// filter in the Ticket Change History preferences, to hide field changes from the history and even exclude entries without a comment For more details see http://trac.edgewall.org/wiki/TracDev/ReleaseNotes/1.0#DetailedUserVisibleC… Cheers, hvr

2 2

RE: [GHC] #7685: :script command does not resolve ~
by Simon Peyton-Jones 09 Oct '13

09 Oct '13

Should there be some documentation in the user manual? (Apols if there already.) Simon | -----Original Message----- | From: ghc-tickets [mailto:ghc-tickets-bounces@haskell.org] On Behalf Of | GHC | Sent: 09 October 2013 09:36 | Cc: ghc-tickets(a)haskell.org | Subject: Re: [GHC] #7685: :script command does not resolve ~ | | #7685: :script command does not resolve ~ | -------------------------------------+---------------------------------- | -- | Reporter: nomeata | Owner: | Type: bug | Status: patch | Priority: normal | Milestone: 7.8.1 | Component: GHCi | Version: 7.6.2 | Resolution: | Keywords: | Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple | Type of failure: None/Unknown | Difficulty: Unknown | Test Case: | Blocked By: | Blocking: | Related Tickets: | -------------------------------------+---------------------------------- | -- | | Comment (by Joachim Breitner <mail@…>): | | In [changeset:"ddc79b229d58c1bed8ae19b955b491436e015e96/ghc"]: | {{{ | #!CommitTicketReference repository="ghc" | revision="ddc79b229d58c1bed8ae19b955b491436e015e96" | GHCi: Expand ~ in :script command | | Fixes: #7685 | }}} | | -- | Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7685#comment:4> | GHC <http://www.haskell.org/ghc/> | The Glasgow Haskell Compiler | _______________________________________________ | ghc-tickets mailing list | ghc-tickets(a)haskell.org | http://www.haskell.org/mailman/listinfo/ghc-tickets

2 1

status of ghc HEAD perf on mac
by Carter Schonwald 08 Oct '13

08 Oct '13

hey all, I just did a perf build of HEAD on my mac (with kazu's no haddocks flag) I did a fast run of the test suite: are all these failures known/being addressed? if you see one that isn't currently accounted for, PLEASE email me back so I can create a ticket for it. Unexpected results from: TEST="cgrun071 T5313 linker_unload cgrun057 static001 T1969 T3064 T4801 objc-hi objcpp-hi TH_spliceE5_prof" OVERALL SUMMARY for test run started at Tue Oct 8 15:37:22 2013 EDT 0:34:11 spent to go through 3799 total tests, which gave rise to 18281 test cases, of which 14821 were skipped 28 had missing libraries 3384 expected passes 34 expected failures 3 caused framework failures 2 unexpected passes 9 unexpected failures Unexpected passes: codeGen/should_run cgrun071 (normal) driver T5313 (normal) Unexpected failures: codeGen/should_run cgrun057 [bad stderr] (profasm) driver static001 [bad stderr] (normal) driver/objc objc-hi [exit code non-0] (normal) driver/objc objcpp-hi [exit code non-0] (normal) perf/compiler T1969 [stat not good enough] (normal) perf/compiler T3064 [stat not good enough] (normal) perf/compiler T4801 [stat too good] (normal) rts linker_unload [bad exit code] (normal) th TH_spliceE5_prof [bad exit code] (normal)

2 1

storage.c
by Simon Peyton-Jones 08 Oct '13

08 Oct '13

I'm getting the error below. I'm fixing it by changing the backslash to a vertical bar. It's in a comment so I guess that's the right thing to do? Dunno how anyone else's builds work! Simon rts/sm/Storage.c: In function 'allocate': rts/sm/Storage.c:725:13: error: multi-line comment [-Werror=comment] cc1: all warnings being treated as errors Microsoft Research Limited (company number 03369488) is registered in England and Wales Registered office is at 21 Station Road, Cambridge, CB1 2FB

3 2

Re: [GHC] #8043: Feature Request : Qualified module exports
by Muhaimin Ahsan 08 Oct '13

08 Oct '13

I'm working on it! :-) On Tuesday, October 8, 2013, GHC wrote: > #8043: Feature Request : Qualified module exports > -------------------------------------+------------------------------------ > Reporter: erikd | Owner: > Type: bug | Status: new > Priority: normal | Milestone: > Component: Compiler | Version: 7.7 > Resolution: | Keywords: > Operating System: Unknown/Multiple | Architecture: Unknown/Multiple > Type of failure: Other | Difficulty: Unknown > Test Case: | Blocked By: > Blocking: | Related Tickets: > -------------------------------------+------------------------------------ > > Comment (by simonmar): > > This may be the new location for that discussion: > http://www.haskell.org/pipermail/libraries/2005-March/003139.html > > Could someone please fix the mailing list links! > > -- > Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8043#comment:7> > GHC <http://www.haskell.org/ghc/> > The Glasgow Haskell Compiler > _______________________________________________ > ghc-tickets mailing list > ghc-tickets(a)haskell.org <javascript:;> > http://www.haskell.org/mailman/listinfo/ghc-tickets > -- Muhaimin (Sent from Gmail Mobile)

1 0

FW: ghc src snapshots
by Simon Peyton-Jones 08 Oct '13

08 Oct '13

Austin (or anyone else) What’s the story about nightly snapshot builds? Does this link with Joachim’s work on Travis? Thanks Simon From: Glasgow-haskell-users [mailto:glasgow-haskell-users-bounces@haskell.org] On Behalf Of Jens Petersen Sent: 07 October 2013 10:40 To: glasgow-haskell-users Subject: ghc src snapshots Are recent source tarball snapshots available somewhere? (I mean more recent than http://darcs.haskell.org/ghcBuilder/uploads/ :) Anyway since I haven't seen any recently, I finally rolled one today and uploaded ghc-7.7.20131005-src.tar.bz2 (11MB) to: http://petersen.fedorapeople.org/ghc/ It seems to build ok for me. :) (I didn't go to the trouble of generating a testsuite tarball too.) Thanks, Jens

2 1

#4012
by Alexander Kjeldaas 08 Oct '13

08 Oct '13

I am confused while reading http://ghc.haskell.org/trac/ghc/ticket/4012<http://ghc.haskell.org/trac/ghc/ticket/8276> . What initially was filed as a difference in results from two identical compiler versions bootstrapped by two different compiler versions seems to morph into a discussion about API stability for differences in code, and various fixes that have been applied. What are the current requirements for getting deterministic builds? Alexander

1 0

New page on the wiki: Emacs tips and tricks
by Jan Stolarek 08 Oct '13

08 Oct '13

Hi all, I created a wiki page that gathers useful Emacs configuration: http://ghc.haskell.org/trac/ghc/wiki/Emacs My intention for this page was to be as self contained as possible, so that you can just copy-paste configuration to your .emacs file and benefit from it instantly (well, after restarting Emacs or reloading config). I'm sure many of you also have interesting configuration and usage tips to share. Richard, on the Newcomers page you mentioned that you have hotkeys for building GHC and searching the codebase - could please tell us how to do this? Janek

3 2

ghc-compete, a repository of fingerprints, and continuous integration
by Joachim Breitner 08 Oct '13

08 Oct '13

Dear Devs, in the best scratch-an-itch-tradition I have created a special git repository, called “ghc-complete”, which tracks the combined state of all GHC-related repositories. It uses GHC’s fingerprint utility and simply tracks the change of the fingerprint file over time: https://github.com/nomeata/ghc-complete The main use of this is to have a repo which I can enable to be continuously checked by Travis (a free CI hoster). So instead of having several people ask on IRC about what build or test suite failures are actually not their fault, they can now simply check the travis page: https://www.travis-ci.org/nomeata/ghc-complete/builds The Travis tests are a bit stripped down, in order to finish reliably before the 50 minute timeout, see https://github.com/nomeata/ghc-complete/blob/master/validate.sh for the precise options. The log files are too large for the JavaScript enhanced view on Travis. So if you click on a specific build result, e.g. https://www.travis-ci.org/nomeata/ghc-complete/builds/12216328 then better immediately browse to the raw build log (the rounded square button in the top right corner with the four horizontal bars). Another use of this repo might be bisecting a problem. If you find this useful, I’d propose to let this run on git.haskell.org proper, and also not as a cronjob (every 15 minutes at the moment), but rather triggered by the actual pushes. On my TODO list is to make the commit messages of the repository more useful, by including the log of the new commits in it. If (not sure if this is planned to that extend) eventually all related repositories of GHC are submodules, this will have become obsolete. Which of course is fine. Enjoy, Joachim -- Joachim “nomeata” Breitner mail(a)joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata(a)joachim-breitner.de • GPG-Key: 0x4743206C Debian Developer: nomeata(a)debian.org

2 1