ghc 7.8.1 has hit [haskell-core]

Magnus Therning

11 Apr 2014 11 Apr '14

2:31 p.m.

This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again. When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Most software today is very much like an Egyptian pyramid with millions of bricks piled on top of each other, with no structural integrity, but just done by brute force and thousands of slaves. -- Alan Kay

Attachments:

attachment.sig (application/pgp-signature — 198 bytes)

Show replies by date

Henrik Johansson

11 Apr 11 Apr

2:44 p.m.

warning: cannot resolve "ghc=7.8.0.20140228-1", a dependency of "haskell-blaze-builder-conduit" warning: cannot resolve "haskell-blaze-builder=0.3.3.2-56", a dependency of "haskell-blaze-builder-conduit" warning: cannot resolve "haskell-conduit=1.0.17.1-2", a dependency of "haskell-blaze-builder-conduit" warning: cannot resolve "haskell-text=1.1.0.1-1", a dependency of "haskell-blaze-builder-conduit" :: The following package cannot be upgraded due to unresolvable dependencies: haskell-blaze-builder-conduit Is this why this happens? On Fri, Apr 11, 2014 at 8:31 PM, Magnus Therning wrote:

...

This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again.

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

/M

-- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus

Most software today is very much like an Egyptian pyramid with millions of bricks piled on top of each other, with no structural integrity, but just done by brute force and thousands of slaves. -- Alan Kay

_______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

Magnus Therning

2:47 p.m.

On Fri, Apr 11, 2014 at 08:44:52PM +0200, Henrik Johansson wrote:

...

warning: cannot resolve "ghc=7.8.0.20140228-1", a dependency of "haskell-blaze-builder-conduit" warning: cannot resolve "haskell-blaze-builder=0.3.3.2-56", a dependency of "haskell-blaze-builder-conduit" warning: cannot resolve "haskell-conduit=1.0.17.1-2", a dependency of "haskell-blaze-builder-conduit" warning: cannot resolve "haskell-text=1.1.0.1-1", a dependency of "haskell-blaze-builder-conduit" :: The following package cannot be upgraded due to unresolvable dependencies: haskell-blaze-builder-conduit

Is this why this happens?

Ah, yes, that particular one is due to blaze-builder-conduit being deprecated in favour of conduit-extra. The same goes for network-conduit. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Goto labels should be left-aligned in all caps and should include the programmer's name, home phone number, and credit card number. -- Abdul Nizar

Linus Arver

12 Apr 12 Apr

6:47 a.m.

On Fri, Apr 11, 2014 at 08:31:48PM +0200, Magnus Therning wrote:

...

This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again.

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

Hello, I noticed that xmonad-contrib is missing. I use this package for my xmonad configuration. -L

Nicola Squartini

11:23 a.m.

And now 7.8.2 is out! On Sat, Apr 12, 2014 at 12:47 PM, Linus Arver wrote:

...

On Fri, Apr 11, 2014 at 08:31:48PM +0200, Magnus Therning wrote:

...
This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again.

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

Hello, I noticed that xmonad-contrib is missing. I use this package for my xmonad configuration.

-L _______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

Magnus Therning

11:55 a.m.

On Sat, Apr 12, 2014 at 05:23:59PM +0200, Nicola Squartini wrote:

...

And now 7.8.2 is out!

Yeah, I know. However, I've been building ghc and the entire repo a few times already these past days so I'm holding off the upgrade to 7.8.2 until tomorrow evening. I hope that doesn't ruin anyone's weekend. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Heuristic is an algorithm in a clown suit. It’s less predictable, it’s more fun, and it comes without a 30-day, money-back guarantee. -- Steve McConnell, Code Complete

Magnus Therning

1:30 p.m.

On Sat, Apr 12, 2014 at 03:47:22AM -0700, Linus Arver wrote:

...

On Fri, Apr 11, 2014 at 08:31:48PM +0200, Magnus Therning wrote:

...
This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again.

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

Hello, I noticed that xmonad-contrib is missing. I use this package for my xmonad configuration.

0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;) /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Any fool can write code that a computer can understand. Good programmers write code that humans can understand. -- Martin Fowler

Sebastian Schwarz

2:09 p.m.

On 2014-12-04, Magnus Therning wrote:

...

0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;)

Is it due to this issue? https://code.google.com/p/xmonad/issues/detail?id=563 Quickly looking for patches for MultiToggle.hs on the XMonad mailing list has turned up this: http://thread.gmane.org/gmane.comp.lang.haskell.xmonad/13649/focus=3186 But I haven't tested it, yet.

Nicola Squartini

3:30 p.m.

I have a problem with HLint after upgrading to 7.8.1. It gives me the error Illegal instruction (core dumped) whenever I run it on a file. I reported the bug https://github.com/ndmitchell/hlint/issues/39 but it seems they are not able to reproduce it. Could someone try it? It might be a problem of this particular binary. On Sat, Apr 12, 2014 at 8:09 PM, Sebastian Schwarz wrote:

...

On 2014-12-04, Magnus Therning wrote:

...
0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;)

Is it due to this issue? https://code.google.com/p/xmonad/issues/detail?id=563

Quickly looking for patches for MultiToggle.hs on the XMonad mailing list has turned up this: http://thread.gmane.org/gmane.comp.lang.haskell.xmonad/13649/focus=3186

But I haven't tested it, yet. _______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

Rudy Matela

4:06 p.m.

The HLint is ok for my after the update: $ hlint Program.hs No suggestions My versions: $ ghc --version The Glorious Glasgow Haskell Compilation System, version 7.8.1 $ hlint --version HLint v1.8.60, (C) Neil Mitchell 2006-2014 $ pacman -Qi haskell-hlint Name : haskell-hlint Version : 1.8.60-2 Architecture : x86_64 $ pacman -Qi ghc Name : ghc Version : 7.8.1-1 Architecture : x86_64 On Sat, Apr 12, 2014 at 8:30 PM, Nicola Squartini wrote:

...

I have a problem with HLint after upgrading to 7.8.1. It gives me the error

Illegal instruction (core dumped)

whenever I run it on a file. I reported the bug https://github.com/ndmitchell/hlint/issues/39 but it seems they are not able to reproduce it. Could someone try it? It might be a problem of this particular binary.

On Sat, Apr 12, 2014 at 8:09 PM, Sebastian Schwarz wrote:

...
On 2014-12-04, Magnus Therning wrote:

...
0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;)

Is it due to this issue? https://code.google.com/p/xmonad/issues/detail?id=563

Quickly looking for patches for MultiToggle.hs on the XMonad mailing list has turned up this: http://thread.gmane.org/gmane.comp.lang.haskell.xmonad/13649/focus=3186

But I haven't tested it, yet. _______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

_______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

Magnus Therning

13 Apr 13 Apr

1:49 a.m.

On Sat, Apr 12, 2014 at 09:06:45PM +0100, Rudy Matela wrote:

...

The HLint is ok for my after the update:

$ hlint Program.hs No suggestions

My versions:

$ ghc --version The Glorious Glasgow Haskell Compilation System, version 7.8.1

$ hlint --version HLint v1.8.60, (C) Neil Mitchell 2006-2014

$ pacman -Qi haskell-hlint Name : haskell-hlint Version : 1.8.60-2 Architecture : x86_64

$ pacman -Qi ghc Name : ghc Version : 7.8.1-1 Architecture : x86_64

It works fine for me too. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Most software today is very much like an Egyptian pyramid with millions of bricks piled on top of each other, with no structural integrity, but just done by brute force and thousands of slaves. -- Alan Kay

Nicola Squartini

7:02 a.m.

Well, this is really weird! I even unistalled all haskell, cleared the pacman cache and reinstalled ghc and hlint, and still I have that "Illegal instruction". So it must be a problem of my system. On Sat, Apr 12, 2014 at 10:06 PM, Rudy Matela wrote:

...

The HLint is ok for my after the update:

$ hlint Program.hs No suggestions

My versions:

$ ghc --version The Glorious Glasgow Haskell Compilation System, version 7.8.1

$ hlint --version HLint v1.8.60, (C) Neil Mitchell 2006-2014

$ pacman -Qi haskell-hlint Name : haskell-hlint Version : 1.8.60-2 Architecture : x86_64

$ pacman -Qi ghc Name : ghc Version : 7.8.1-1 Architecture : x86_64

On Sat, Apr 12, 2014 at 8:30 PM, Nicola Squartini wrote:

...
I have a problem with HLint after upgrading to 7.8.1. It gives me the error

Illegal instruction (core dumped)

whenever I run it on a file. I reported the bug https://github.com/ndmitchell/hlint/issues/39 but it seems they are not able to reproduce it. Could someone try it? It might be a problem of this particular binary.

On Sat, Apr 12, 2014 at 8:09 PM, Sebastian Schwarz wrote:

...
On 2014-12-04, Magnus Therning wrote:

...
0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;)

Is it due to this issue? https://code.google.com/p/xmonad/issues/detail?id=563

Quickly looking for patches for MultiToggle.hs on the XMonad mailing list has turned up this: http://thread.gmane.org/gmane.comp.lang.haskell.xmonad/13649/focus=3186

But I haven't tested it, yet. _______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

_______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

Bjørn Øivind Bjørnsen

7:05 a.m.

On Sunday 13 April 2014 13:02:15 Nicola Squartini wrote:

...

Well, this is really weird! I even unistalled all haskell, cleared the pacman cache and reinstalled ghc and hlint, and still I have that "Illegal instruction". So it must be a problem of my system.

It may be a long shot, but have you tried dropping the kernel caches? Perhaps you have a bad stick of RAM or something. echo 3 > /proc/sys/vm/drop_caches as root should do the trick. You may also want to do a md5sum of the hlint binary before and after you do the above, to see if it makes a difference. Best Regards, Bjørn Øivind

Nicola Squartini

8:41 a.m.

I tried on another system and it works fine. It's definitely a problem of this machine, but not a memory corruption because I have ecc and edac-util shows 0 error corrections. On Sun, Apr 13, 2014 at 1:05 PM, Bjørn Øivind Bjørnsen < bo.bjornsen@gmail.com> wrote:

...

On Sunday 13 April 2014 13:02:15 Nicola Squartini wrote:

...
Well, this is really weird! I even unistalled all haskell, cleared the pacman cache and reinstalled ghc and hlint, and still I have that "Illegal instruction". So it must be a problem of my system.

It may be a long shot, but have you tried dropping the kernel caches? Perhaps you have a bad stick of RAM or something.

echo 3 > /proc/sys/vm/drop_caches

as root should do the trick. You may also want to do a md5sum of the hlint binary before and after you do the above, to see if it makes a difference.

Best Regards, Bjørn Øivind _______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

Linus Arver

3:18 a.m.

On Sat, Apr 12, 2014 at 07:30:02PM +0200, Magnus Therning wrote:

...

On Sat, Apr 12, 2014 at 03:47:22AM -0700, Linus Arver wrote:

...
On Fri, Apr 11, 2014 at 08:31:48PM +0200, Magnus Therning wrote:

...
This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again.

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

Hello, I noticed that xmonad-contrib is missing. I use this package for my xmonad configuration.

0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;)

Well, I tried a simple 1-liner "patch" (see attachment) --- it removes the type signature from XMonad/Layout/MultiToggle.hs at line 162. I've tried out the patch with cblrepo and everything compiles (quite amazing, as there are over 200 modules to compile and this is the only thing holding back the package) and I can use XMonad again. I don't use the MultiToggle module though, so I cannot say that it's completely stable. -L

...

/M

-- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus

Any fool can write code that a computer can understand. Good programmers write code that humans can understand. -- Martin Fowler

Magnus Therning

1:52 a.m.

On Sun, Apr 13, 2014 at 12:18:12AM -0700, Linus Arver wrote:

...

On Sat, Apr 12, 2014 at 07:30:02PM +0200, Magnus Therning wrote:

...
On Sat, Apr 12, 2014 at 03:47:22AM -0700, Linus Arver wrote:

...
On Fri, Apr 11, 2014 at 08:31:48PM +0200, Magnus Therning wrote:

...
This morning I pushed out ghc 7.8.1 to [haskell-core]. This means that [haskell-testing] will be retired in a few days so please point pacman back to the core repo again.

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

Hello, I noticed that xmonad-contrib is missing. I use this package for my xmonad configuration.

0.11.2 doesn't build out of the box at the moment so it'll likely take a little while until it gets added again. Unless I receive a patch that fixes it of course ;)

Well, I tried a simple 1-liner "patch" (see attachment) --- it removes the type signature from XMonad/Layout/MultiToggle.hs at line 162. I've tried out the patch with cblrepo and everything compiles (quite amazing, as there are over 200 modules to compile and this is the only thing holding back the package) and I can use XMonad again. I don't use the MultiToggle module though, so I cannot say that it's completely stable.

Sometimes that's all it takes. Thanks for the patch. It's building as I write this and should hit the repos in a matter of minutes. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus What gets measured, gets done. -- Tom Peters

Sergey Khorev

19 Apr 19 Apr

8:20 a.m.

...

When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

The upgrade was really smooth. I am wondering if you are going to restore hoogle? I was able to build it in cabal sandbox so hopefully there are no major blockers.

Magnus Therning

20 Apr 20 Apr

2:41 a.m.

On Sat, Apr 19, 2014 at 12:20:26PM +0000, Sergey Khorev wrote:

...

...
When I first created the branch for ghc 7.8 there were a number of packages that weren't buildable, and hence they were removed. I've since re-added a few of them, but not all. Not yet. If you find something crucial to be missing, please let me know so I can fix it post-haste.

The upgrade was really smooth. I am wondering if you are going to restore hoogle? I was able to build it in cabal sandbox so hopefully there are no major blockers.

It's being uploaded at the moment. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus Any fool can write code that a computer can understand. Good programmers write code that humans can understand. -- Martin Fowler