GHC Weekly News - 2015/10/16

[Austin Seipp]

(This post is available online at
https://ghc.haskell.org/trac/ghc/blog/weekly20151016)

Hi *,

Welcome for the latest entry in the GHC Weekly News, which has been
somewhat irregularly scheduled - but we're as busy as ever!

== 8.0.1 release roadmap ==

We're still steaming ahead on GHC 8.0.1 - any interested participants
are suggested to look at the wiki:Status/GHC-8.0.1 page, to see where
we're currently at.

These past few weeks have been good: we've gotten the first part of
the overloaded record fields work in, and we have plans to land the
kind equalities work in November. Furthermore, Simon Marlow, Peter
Wortmann and Ben are working on reviewing all of the DWARF
improvements, and hopefully the start of this work will land next
week.

But 8.0 isn't the only thing that'll be getting released...

== And some other minor releases ==

In a total, unprecedented upset - we're aiming to do ''three'' GHC
releases in a fairly short amount of time.

=== 7.10.3 ===

Ben Gamari has been working on steadily hacking away at GHC 7.10.3,
and the hopes are that we'll be able to ship it soon. This will fix
several dozen bugs, some of which are critical for our users.

You can keep up to date by following the wiki:Status/GHC-7.10.3 page.

=== 7.8.5 ===

Simultaneously, your author will ''also'' be working on a GHC 7.8.5
release. While we were alerted a few months ago to this breakage, it
seems rather unfortunate for the 7.8 series to remain broken on such a
popular OS.

Furthermore, the "Three GHC Release Policy" for many authors - to
support the last three major versions of GHC - would mean that 7.8
would be broken for OS X developers for an ''entire year until GHC
8.2.1''. Which is a pretty unfortunate circumstance.

It's not expected the 7.8.5 release will contain any other fixes, however.


== List chatter ==

(Over the past two weeks)

  - Ben Gamari wrote in about '''switching the users guide to
reStructuredText''', and the TL;DR is - it's done! We'll have a
beautiful new users guide for GHC 8.0.1
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010029.html

  - Matthew Pickering comments about the state of pattern synonym
signatures, remarking that they're currently confusing, noting down
some things we could possibly fix.
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010024.html

  - Ben Gamari talked about the status of the recent DWARF work, and
so far it's looking good. The needed patches are still in the review
pipeline, but the hope is that they'll all be done in time for 8.0.1.
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010039.html

  - David A Roberts wrote in to ask a question: now that we have
`ApplicativeDo`, what about `Applicative` comprehensions? The
responses indicate this seems like it would be a great addition.
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010062.html

  - Richard Eisenberg sent in a status update about his work on kind
equalities, and the hope is it will land shortly in November! (Your
editor then hassled him for a syntax change before landing.)
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010077.html

  - Adam Foltzer has requested another release of the GHC 7.8 series,
due to it being completely broken on OS X El Capitan. Expect more news
on this soon. 
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010078.html

  - Erik de Castro Lopo has recently been working with LLVM, and
decided to publish his automation so interested GHC hackers could keep
up to date and try new LLVMs easily.
https://mail.haskell.org/pipermail/ghc-devs/2015-October/010086.html

== Noteworthy commits ==

(Over the past two weeks)

  - Commit 4fd6207ec6960c429e6a1bcbe0282f625010f52a - Move the users
guide to reStructured Text.

  - Commit 6cde981a8788b225819be28659caddc35b77972d - Make
`GHC.Generics` capable of handling unboxed tuples.

  - Commit 0eb8fcd94b29ee9997b386e64203037bdf2aaa04 - Enable
`Enumeration is empty` warnings for `Integer`

  - Commit 620fc6f909cd6e51b5613454097ec1c9f323839a - Make Windows
linker more robust to unknown sections

  - Commit 5d841108acef950fed6a5e608ac9b18e7431aa87 - Add short
library names support to the Windows linker

  - Commit f8fbf385b879fe177409a25cc9499275ea3dc45d - Reinstate
monomorphism-restriction warnings

  - Commit dcc342870b4d8a739ccbed3ae26e84dcc3579914 - Don't
inline/apply other rules when simplifying a rule RHS.

  - Commit dec5cd4085488686b5ed50bb26ccbc0ba7b645ec - base: Add
`forkOSWithUnmask`

  - Commit e8c8173923302268ef950c3b21e276779e45ac83 - Allow arr ∧
(first ∨ (***)) as minimal definition of Arrow instance

  - Commit 29310b622801733e1b29a9a61988406872db13ca - Switch to LLVM version 3.7

  - Commit 04e8366608fee4f5e3358acc855bc6f556c3f508 - ELF/x86_64: map
object file sections separately into the low 2GB

  - Commit b1884b0e62f62e3c0859515c4137124ab0c9560e - Implement
`DuplicateRecordFields`

  - Commit 75492e7467ff962f2f2e29e5c8b2c588c94ae8a7 - Add typed holes
support in Template 

dns issues with ghc.haskell.org

[lennart spitzner]

hi,

This mail is intended for the maintainers of ghc.haskell.org who i can
hopefully reach on this list.

Once more i run into the issue that firefox resolves ghc.haskell.org
to the wrong (an outdated?) ip address (resulting in some 500 Internal
Server Error page; i guess on some other server..). When switching to
a different browser the address is resolved "correctly" and i reach
the expected server.

For me personally this is only slightly annoying, but another fellow
on irc recently reported the same symptoms, so i am not the only one
affected. Might be worth resolving.

I have not observed a failure of this kind for any other server. I
don't really know what could be the issue or how to diagnose this
further. If it helps, in this case (have not checked when this
occurred previously):
"wrong" address 162.242.239.16
"correct"   104.25.122.13


Lennart
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: "Excuse me, I think this i my stop..." - Resigning from the Platform

[Bardur Arantsson]

On 10/13/2015 05:08 AM, Mark Lentczner wrote:
> I think this is the right time for me to exit:
> 

I'm pretty sure that there are many things that we could agree or
disagree on, but *THANK YOU* for your efforts on improving the Haskell
ecosystem and your efforts to spread the word!

Regards,

___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: Request to reopen the issue of 7.8 on El Capitan

[Austin Seipp]

Thank you Mark! I was going to ask you since I know you have like 6 OS
X virtual machines, so you have much better coverage. :) I'm getting
my hands on a mac now to test it, and I'll let everyone know when it's
ready (I have to do the other annoying stuff of like making the proper
bug and a milestone, etc.)

On Fri, Oct 16, 2015 at 12:33 AM, Mark Lentczner
 wrote:
> I can build a 7.8.5 binary for release if someone can put the patch in the
> GHC tree. (Having a soft spot in my heart for the 7.8 line!)
> I'm leaving on a short trip, and will be back on Tuesday and build middle of
> next week.
> - Mark



-- 
Regards,

Austin Seipp, Haskell Consultant
Well-Typed LLP, http://www.well-typed.com/
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: MonadFail decisions

[David Luposchainsky]

On 16.10.2015 18:41, Edward Kmett wrote:
> As for the "mob"

I'd like to apologize for using that word. I think "crowd" would have been much
closer to what I meant to say. There's no mob to be found here, since all the
opinions - in favour or against - were stated in a very constructive way. So
again, sorry about that.

David
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: MonadFail decisions

[Edward Kmett]

Not a bad idea. I think Herbert was talking about calling it -Wcompat or
something.

On Fri, Oct 16, 2015 at 1:06 PM, Howard B. Golden  wrote:

> On Friday, October 16, 2015 9:22 AM, Edward Kmett wrote:
>
>
> > It sounds like we'll need to delay the warnings themselves until around
> 8.4.
>
> I propose an optional generic flag -fearly-warning (pun slightly intended)
> to get _all_ warnings of planned changes.
>
> Howard
>
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: MonadFail decisions

[Howard B. Golden]

On Friday, October 16, 2015 9:22 AM, Edward Kmett wrote:


> It sounds like we'll need to delay the warnings themselves until around 8.4.

I propose an optional generic flag -fearly-warning (pun slightly intended) to 
get _all_ warnings of planned changes.

Howard
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: MonadFail decisions

[Edward Kmett]

Hi David,

I took the time to update the MonadFail wiki page to include both the
timeline currently under consideration, lengthening the timeline to finish
in 8.8 to comply with the "3 release policy" and to ensure that folks
always have a notification of pending breaking changes.

I included a couple of personal comments about the desugaring in 1.3 where
we could do better. The improvements in 1.3 could be made any time over the
8.0 and 8.2 releases before we start expecting people to cut over in 8.4
without impact.

As for the "mob", please keep in mind that the vast majority of feedback
about the MonadFail proposal has been positive and draw heart from that.
Many of the folks who were against the Foldable/Traversable generalizations
(e.g. Lennart) are heavily in favor of MFP.

-Edward






On Fri, Oct 16, 2015 at 12:09 PM, David Luposchainsky <
dluposchain...@googlemail.com> wrote:

> On 13.10.2015 16:29, Simon Peyton Jones wrote:
> > Also, David, did our conversation at HX help you get un-stuck?
>
> Hi Simon,
>
> yes, it was definitely a good pointer. On the other hand, I found the
> Haskell
> Exchange to be quite a frustrating event with respect to current events:
> there
> was a load of very loud, but in my opinion very wrong, categorical
> opposition to
> breaking changes in general.
> I spent quite a bit of time worrying about MonadFail in the past, but
> right now
> I'd like to wait for a "tentative yes" from the CLC before I keep going,
> because
> I'm really not sure the mob is going to make me throw away my patch.
> Granted, a
> lot of the discussion is about MRP, but many of the points brought up
> there are
> equally valid against the MFP.
>
> David
>
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: MonadFail decisions

[Edward Kmett]

The current intention is to go ahead with MonadFail.

It sounds like we'll need to delay the warnings themselves until around 8.4.

We can add them, but not turn them on by default in the short term. This
has the knock-on effect of delaying the whole plan a release or two, but
otherwise the plan is very actionable.

A lot of the opposition comes from fear that we 'might do anything at any
time'. If we're up front about what is coming and give sufficient notice
and the ability for folks to maintain a reasonably wide backwards
compatibility window without needing to dip into CPP or suppress warnings
them most of those fears go away.

-Edward

On Fri, Oct 16, 2015 at 12:09 PM, David Luposchainsky <
dluposchain...@googlemail.com> wrote:

> On 13.10.2015 16:29, Simon Peyton Jones wrote:
> > Also, David, did our conversation at HX help you get un-stuck?
>
> Hi Simon,
>
> yes, it was definitely a good pointer. On the other hand, I found the
> Haskell
> Exchange to be quite a frustrating event with respect to current events:
> there
> was a load of very loud, but in my opinion very wrong, categorical
> opposition to
> breaking changes in general.
> I spent quite a bit of time worrying about MonadFail in the past, but
> right now
> I'd like to wait for a "tentative yes" from the CLC before I keep going,
> because
> I'm really not sure the mob is going to make me throw away my patch.
> Granted, a
> lot of the discussion is about MRP, but many of the points brought up
> there are
> equally valid against the MFP.
>
> David
>
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Re: MonadFail decisions

[David Luposchainsky]

On 13.10.2015 16:29, Simon Peyton Jones wrote:
> Also, David, did our conversation at HX help you get un-stuck?

Hi Simon,

yes, it was definitely a good pointer. On the other hand, I found the Haskell
Exchange to be quite a frustrating event with respect to current events: there
was a load of very loud, but in my opinion very wrong, categorical opposition to
breaking changes in general.
I spent quite a bit of time worrying about MonadFail in the past, but right now
I'd like to wait for a "tentative yes" from the CLC before I keep going, because
I'm really not sure the mob is going to make me throw away my patch. Granted, a
lot of the discussion is about MRP, but many of the points brought up there are
equally valid against the MFP.

David
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

RE: [Differential] [Commented On] D761: Implement DuplicateRecordFields

[Simon Peyton Jones]

Adam

| Is it okay for me to go ahead and merge this? I'll resolve
|  whatever conflicts there are and do a final validate before merging.I think 
so, yes.

You might want to talk to Matthew Pickering, who is scared about the rebase 
work he'll have to do on https://phabricator.haskell.org/D1152

Simon


|  -Original Message-
|  From: nore...@phabricator.haskell.org
|  [mailto:nore...@phabricator.haskell.org]
|  Sent: 16 October 2015 09:13
|  To: Simon Peyton Jones
|  Subject: [Differential] [Commented On] D761: Implement DuplicateRecordFields
|  
|  adamgundry added a comment.
|  
|  This builds fine on Travis
|   but not on
|  Phab, I suspect because the diff needs to be updated to apply cleanly to the
|  latest HEAD. Is it okay for me to go ahead and merge this? I'll resolve
|  whatever conflicts there are and do a final validate before merging.
|  
|  
|  REPOSITORY
|rGHC Glasgow Haskell Compiler
|  
|  REVISION DETAIL
|https://phabricator.haskell.org/D761
|  
|  EMAIL PREFERENCES
|https://phabricator.haskell.org/settings/panel/emailpreferences/
|  
|  To: adamgundry, goldfire, bgamari, simonpj, austin
|  Cc: sjcjoosten, haggholm, mpickering, bgamari, tibbe, thomie, goldfire
___
ghc-devs mailing list
ghc-devs@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs