Bug#850915: ghc: Please switch to llvm 3.8 or, better, 3.9

[Gianfranco Costamagna]

On Mon, 19 Jun 2017 13:31:33 + (UTC) Gianfranco Costamagna 
<locutusofb...@debian.org> wrote:
> Hello,
> >Now that we shipped strech, I really would like to see llvm 3.7 removed.
> >Can we go ahead now ?
> 
> 
> as said, with ghc 8.0.1 and 8.0.2 this seems a little impossible, but as 
> Clint said on debian-haskell
> 
> >> last thing: should we switch llvm? wrt #850915
> >I think the GHC 8.2 release candidates are targeting LLVM 4.0.
> 
> 
> so for Buster we will probably ship 8.2 or something even newer.
> 
> But for now, I don't think such change is worth the effort, specially because 
> it will probably introduce
> bugs because this will be an untested configuration.
> I really hope with 8.2 to make this one disappear :)
> (I guess it will be released in some months)
> 

3.7 can now be removed, because ghc 8.2 (using 3.9) is now in unstable and 
built.
I also uploaded ghc 8.4 that uses llvm 5.0, so maybe we will move forward in a 
few weeks with this one.

For llvm 6.0, I'm sorry but it won't be probably be a real plan for one year or 
so...

I hope this is enough for now, I see 3.8,3.9 are blocked by other 
reverse-dependencies and 5.0 is fairly new now.

Let me know, or close this bug if you think this is ok for now!

Gianfranco

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Accepted xmonad 0.13-7 (source) into unstable

[Gianfranco Costamagna]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Format: 1.8
Date: Tue, 13 Mar 2018 16:38:07 +0100
Source: xmonad
Binary: xmonad libghc-xmonad-dev libghc-xmonad-prof libghc-xmonad-doc
Architecture: source
Version: 0.13-7
Distribution: unstable
Urgency: medium
Maintainer: Debian Haskell Group 
<pkg-haskell-maintainers@lists.alioth.debian.org>
Changed-By: Gianfranco Costamagna <locutusofb...@debian.org>
Description:
 libghc-xmonad-dev - Lightweight X11 window manager${haskell:ShortBlurb}
 libghc-xmonad-doc - Lightweight X11 window manager${haskell:ShortBlurb}
 libghc-xmonad-prof - Lightweight X11 window manager${haskell:ShortBlurb}
 xmonad - Lightweight X11 window manager written in Haskell
Closes: 892004
Changes:
 xmonad (0.13-7) unstable; urgency=medium
 .
   [ Ilias Tsitsimpis ]
   * Change Priority to optional. Since Debian Policy version 4.0.1,
 priority extra has been deprecated.
   * Use the HTTPS form of the copyright-format URL
   * Modify d/watch and Source field in d/copyright to use HTTPS
   * Declare compliance with Debian policy 4.1.1
   * Use salsa.debian.org URLs in Vcs-{Browser,Git} fields
 .
   [ Gianfranco Costamagna ]
   * From Jeremy Bicha: adapt to gnome-settings-daemon 3.28
 (Closes: #892004)
Checksums-Sha1:
 e0e79fec6e5349cdb5e7aaff61687806a5205924 2851 xmonad_0.13-7.dsc
 50292503aa8cc411ef8a2f7520631d5e29e47611 13600 xmonad_0.13-7.debian.tar.xz
 7886bde236301558c37e2b47084fc729b022509c 9332 xmonad_0.13-7_source.buildinfo
Checksums-Sha256:
 ad0b173722eedb68047a89d0dd6edcf73d947b7044b9ea8bcde92e4b60ce006e 2851 
xmonad_0.13-7.dsc
 5278e518db9b5d2da68293e5f7fce0d609b0a35fd49dda2dc02c69ab700fc6f3 13600 
xmonad_0.13-7.debian.tar.xz
 67007a0a70f1ede8bb5973ec931aedfcb8bd0c004ebe1cb659107778ba2a8091 9332 
xmonad_0.13-7_source.buildinfo
Files:
 f47a4c3e51355acea1fe59465efc8b8b 2851 haskell optional xmonad_0.13-7.dsc
 d9188e7940ca921749ea0b873982226c 13600 haskell optional 
xmonad_0.13-7.debian.tar.xz
 899abb331244aac79a0eb54433121c87 9332 haskell optional 
xmonad_0.13-7_source.buildinfo

-BEGIN PGP SIGNATURE-
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=mjNF
-END PGP SIGNATURE-


___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#892004: Please adapt to gnome-settings-daemon 3.28 by March 15

[Gianfranco Costamagna]

hello,

>I am unable to do something this week. Could someone else apply the
>change and upload the package?


in deferred/2

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Re: How to print test log in case of failures?

[Gianfranco Costamagna]

Hello James,


>That whole file looks like a major source of unaligned accesses, at aglance, 
>and likely needs some attention to support a potential lack of 64-bit
>(and, in general, any) unaligned accesses.


so, thanks for the really helpful answer!

Do you think it is ok to override the testsuite on armhf, and assume that 
people will not mix arm64 and armhf at runtime?
I don't think I have the knowledge to understand or fix it... I think disabling 
is the best thing we can do for now...

what is your opinion?
(I still see no big usage of haskell on arm...)

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

How to print test log in case of failures?

[Gianfranco Costamagna]

Hello, I'm trying to sort out an armhf alignment issue
that happens with haskell-th-utilities in Ubuntu
(but I'm pretty sure this isn't an Ubuntu-specific issue, reason
for Ubuntu to fail is probably because stuff is run on top of an arm64
kernel).

Test suite logged to: dist-ghc/test/th-utilities-0.2.0.1-test.log
0 of 1 test suites (0 of 1 test cases) passed.
/usr/share/cdbs/1/class/hlibrary.mk:154: recipe for target 'check-ghc-stamp' 
failed
make: *** [check-ghc-stamp] Error 1

this is what is printed.

Now, how can we print that file in case of failures?
I tried an override_dh_auto_test but it seems to be a cdbs packaging, and I 
don't know
if haskell helpers have some "hooks" to intercept failures
(it would be even better to patch the build system to automatically print the 
content
of the testsuite log in case of failures BTW)

If you are interested, this is the log
https://launchpad.net/ubuntu/+source/haskell-th-utilities/0.2.0.1-1build1/+build/13678151

Any help is appreciated :)

(I'm ccing James, since he knows a lot on this aligment/topic :) )

thanks!

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Re: Bug#853285: closed by Clint Adams <cl...@debian.org> (Bug#853285: fixed in ghc 8.0.2-3)

[Gianfranco Costamagna]

Hello,


>Here's an updated patch for 8.0.2-9.


committed (I don't even know how to test it properly) on git, but obviously not 
uploaded.
(I don't think it is worth an upload)

Please forward the patches upstream whenever possible!
thanks

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#867853: gcc: error: unrecognized command line option ‘-no-pie’

[Gianfranco Costamagna]

control: forcemerge 871740 867853

> Okay, I'll take this to mean that if I submit a working/useful patch, you'll 
> apply it.  I'll get to work on that.

consider that: gcc-5 is not part of any Debian stable release, and gcc-4.9 is 
has been part of jessie for a while, but
jessie was released already with a pie one.
So, this is in fact affecting no stable or unstable release.

We might add a versioned conflict, and drop it after some months, even if I 
don't know how much we will solve with that
(also the opposite won't probably work, older ghc and newer gcc

G.




signature.asc
Description: OpenPGP digital signature
___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#871740: gcc: error: unrecognized command line option ‘-no-pie’

[Gianfranco Costamagna]

control: forcemerge -1 867853
control: unarchive 867853
control: reopen 867853
control: tags 867853 moreinfo

> Okay, I'll take this to mean that if I submit a working/useful patch, you'll 
> apply it.  I'll get to work on that.

consider that: gcc-5 is not part of any Debian stable release, and gcc-4.9 is 
has been part of jessie for a while, but
jessie was released already with a pie one.
So, this is in fact affecting no stable or unstable release.

We might add a versioned conflict, and drop it after some months, even if I 
don't know how much we will solve with that
(also the opposite won't probably work, older ghc and newer gcc

G.



signature.asc
Description: OpenPGP digital signature
___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

[no subject]

[Gianfranco Costamagna]

salutations 



http://szbttx.com/editproduct.php?pair=274cv7a3xdn



Thanks so much
Gianfranco
___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#867853: closed by Gianfranco Costamagna <locutusofb...@debian.org> (Re: Bug#867853: gcc: error: unrecognized command line option ‘-no-pie’)

[Gianfranco Costamagna]

Hello,




>I'm not sure I understand this response.
>
>When you say "gcc-specific", are you saying that the Debian Haskell package 
>supports compilers other than gcc?  It appears to me that it specifically 
>requires gcc.


I mean, that the flag has been added by gcc, and a lot of programs in the 
archive are supplying it to the build, with a backward incompatible
change.

e.g. you can think about all the -dkms packages, because pie is something that 
the kernel can't use.

So, instead of patching every single control file of a tool using no-pie, I 
would suggest to patch gcc if there still are supported configurations
without the -f-no-pie switch.
(I mean, implement that switch and do nothing when used)

>When you say "fix the whole archive" what do you mean?


the above :)
>I suggest only a update to the "debian/control" file to prevent/discourage 
>users from trying to use ghc with older, unsupported compilers.


updating control file, adding version constraints with versions not in the 
archive anymore is kind of useless, speaking about Buster we won't have such
gcc versions anymore since at least two stable releases.
>Last question:

>If I contribute a patch to the debian/control file to do this, would it get 
>applied?

nobody complained so far about this issue, so you are probably one of the few 
people not updating their compiler :)

I'm really not sure this is worth the effort

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#865711: haskell-yaml: FTBFS everywhere (except all)

[Gianfranco Costamagna]

Hello,

>I don't know what we should do here.  The behavior of the embedded
>libyaml differs from our libyaml and I'm not sure why, but it appears

>to be on purpose.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=865675
because the embedded yaml is more up-to-date wrt the Debian one, and for
this reason I opened a bug report.

Patching yaml in Ubuntu made the whole stuff migrate, however seems that
some unintended side-effects are possible, according to pyyaml regressions
on i386.

https://github.com/snoyberg/yaml/issues/110

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#850915: ghc: Please switch to llvm 3.8 or, better, 3.9

[Gianfranco Costamagna]

Hello,
>Now that we shipped strech, I really would like to see llvm 3.7 removed.
>Can we go ahead now ?


as said, with ghc 8.0.1 and 8.0.2 this seems a little impossible, but as Clint 
said on debian-haskell

>> last thing: should we switch llvm? wrt #850915
>I think the GHC 8.2 release candidates are targeting LLVM 4.0.


so for Buster we will probably ship 8.2 or something even newer.

But for now, I don't think such change is worth the effort, specially because 
it will probably introduce
bugs because this will be an untested configuration.
I really hope with 8.2 to make this one disappear :)
(I guess it will be released in some months)

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#855627: ghc-doc: ships dangling symlink /usr/share/doc/ghc-doc/html/users_guide/prof_scc -> prof_scc.png

[Gianfranco Costamagna]

control: tags -1 pending
>during a test with piuparts I noticed your package ships a dangling

>symlink:


Hi,

committed on git master, not sure if worth an upload (I guess not)

https://anonscm.debian.org/cgit/pkg-haskell/DHG_packages.git/commit/?id=be033b6558f51ae50b60c051e70ce3397d92f029

thanks

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#853285: ghc: Patch to add support for cross-compilation of GHC

[Gianfranco Costamagna]

Hello:



>* ghc does have a test suite, but the Debian package doesn't run it.


ghc-testsuite is the source package "running" it.

(please don't ask me the reasons for the split :) )

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#850920: Cabal-debian bundled packages

[Gianfranco Costamagna]

Hello Kei,
On Tue, 17 Jan 2017 19:42:57 + David Fox  wrote:
> Fortunately or unfortunately, I made extensive changes to this code in
> November:
> 

I did apply your patch and the head result didn't change

> https://github.com/ddssff/cabal-debian/commit/480f4f099657a20eb46a45c0ca5f492038773e5b
> 
> Could you test the latest version of cabal-debian and see if it resolves
> your issue?  Thanks!
this version returns the following (in a clean sid chroot)

head -12 debian/control
Source: haskell-time-locale-compat
Maintainer: Debian Haskell Group 

Priority: extra
Section: haskell
Build-Depends: debhelper (>= 10),
 haskell-devscripts-minimal | haskell-devscripts (>= 0.8),
 cdbs,
 ghc,
 ghc-prof,
 libghc-old-locale-dev,
 libghc-old-locale-prof,
Build-Depends-Indep: ghc-doc,

is that good?

package is available here [1]

[1] 
http://debomatic-amd64.debian.net/distribution#unstable/cabal-debian/4.35.6-1/lintian

G.




signature.asc
Description: OpenPGP digital signature
___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#850915: ghc: Please switch to llvm 3.8 or, better, 3.9

[Gianfranco Costamagna]

Hello

>Probably need to backport this upstream commit:

>
>https://git.haskell.org/ghc.git/commitdiff/672314cbeb8ac386a58f17dc4650dbdf4c55d8b5


probably not only that one, but at least 3 commits

Unfortunately when I expressed the idea (and the commits) over irc on 
#debian-haskell or whatever
I got a nack, because diverging from upstream in the default toolchain version 
is a no-go,
and moreover it would require probably a lot of binNMUs in the arm* 
architectures for the whole
haskell stack.

Upstream can fix bugs when llvm 3.7 is used, now when we diverge in such 
important key packages.

(this is a sum of the discussion I recall right now, I can search irc logs if 
needed)

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

llvm-toolchain-3.8 needs fix on armel, or ghc will be removed there.

[Gianfranco Costamagna]

Hi,

simple as it.

We would like to have only llvm-toolchain-3.8 and llvm-toolchain-3.9 in Stretch.

So, in case nobody has a patch/opinion on making llvm-3.8+ build/work on armel,
I'm afraid no haskell packages will be available there for Stretch.

Simple as this, keeping llvm-3.7 just for armel architecture seems an overkill,
we already removed armel once, and probably this will happen again.

You can see the bug report here [1] where some solutions are provided.

thanks!

[1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=820535


Gianfranco

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Bug#712228: Hardening flag -pie breaks compilation with GHC

[Gianfranco Costamagna]

Hi,


>With this change the package builds fine in unstable and experimental,
>thank you.


oops fixed :)

G.

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers

Re: Finding sponsor

[Gianfranco Costamagna]

yes,
Debian Haskell Group (in cc) might give you some hints :)

G.




Il Giovedì 2 Giugno 2016 14:47, Marwan Rabbâa  ha scritto:



but still have a question

this app is in haskell and I built i with stack
https://www.stackage.org

I do not thing stack use internal distribution libraries.

Is there a maintainer of an haskell package that could answer ?



2016-06-02 12:14 GMT+02:00 Marwan Rabbâa :

ok, so I'll delay the first step is to make a production package for my use)
>
>thanks
>
>
>
>2016-06-02 11:52 GMT+02:00 Andrew Shadura :
>
>On 2 June 2016 at 11:51, Marwan Rabbâa  wrote:
>>> My package is
>>> https://packagecloud.io/waghanza/plunder/packages/debian/jessie/postgrest_0.3.1_amd64.deb
>>> currently
>>>
>>> I support only binary form (I do not support .dsc).
>>>
>>> Can I offer my package community nosw or should I wait 'till I support to
>>> create dsc files ?
>>
>>No, you need to make a source package to be able to upload it to Debian.
>>
>>--
>>Cheers,
>>  Andrew
>>
>

___
Pkg-haskell-maintainers mailing list
Pkg-haskell-maintainers@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-haskell-maintainers