[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: GHC packaging collaboration



Hi Jens,

Am Dienstag, den 10.01.2012, 02:16 -0500 schrieb Jens Petersen:
> > You can find our patches against the ghc package in
> > http://anonscm.debian.org/darcs/pkg-haskell/ghc/patches/
> > and for the experimental package (7.4.1-rc1) in
> > http://anonscm.debian.org/darcs/pkg-haskell/experimental/ghc/patches/
> > 
> > Especially the system-libffi-patch is used by both of us, and is
> > always a pain to port to new versions. We could try to generalize it so that
> > it can go into upstream proper and be enabled by a configure flag.
> > Anybody (you, or someone from the Debian Haskell Group) interested in this
> > task?
> 
> Right - perhaps I can take a stab at it one day if noone else has time.
> 
> Yes this patch is probably the main maintenance burden currently for ghc
> so it would definitely be good to upstream - perhaps it needs a switch
> in configure though or Linux should just use system libffi anyway.
> (Though earlier when I proposed the latter to SimonM, he was not too keen.)

I think a switch would definitely be more likely accepted.

> > Judging from the contents of the patch
> > ghc-gen_contents_index-type-level.patch you want to avoid the large
> > runtime of gen_contents_index when type-level is installed. I had
> > send a
> > patch upstream that improves on that, maybe you can drop the patch
> > when
> > packaging ghc 7.4.1:
> > https://github.com/ghc/haddock/commit/b50695a86358fd02b9a0ececbc107bf5274d5f95
> 
> I see thanks.  Ok I will definitely consider that.
> So with that you don't see any lag with installing haddock documentation
> any more, or probably debian spawns a subprocess for re-indexing anyway?

Actually, I have not tried it yet, we only test-built 7.4.1, but not put
in in the main repository (sid a.k.a. unstable) :-)

In Debian, the re-indexing does not happen in the background, but thanks
to dpkg triggers, it happens at most once per installation, even if you
install multiple Haskell packages; if in Fedora it is re-indexed per
newly installed package I understand that the lag is more sever.

> > The 7.4.1-rc1 release had build failures on almost every architecture
> > besides i386/amd64. I guess some of these will affect you as well, so
> > any help is appreciated:
> > http://hackage.haskell.org/trac/ghc/ticket/5733
> > http://hackage.haskell.org/trac/ghc/ticket/5735
> > http://hackage.haskell.org/trac/ghc/ticket/5732
> > http://hackage.haskell.org/trac/ghc/ticket/5734
> 
> Ok, I will try to do a testbuild of rc1 for Fedora finally soon too.

All these bugs are sorted out by now, and the package builds fine on all
arches:
https://buildd.debian.org/status/package.php?p=ghc&suite=experimental

> > BTW, if you are curious, you are welcome to subscribe to
> > debian-haskell@lists.debian.org at
> > http://lists.debian.org/debian-haskell/. Also, if you want to contact
> > us, that is the right address, even if you are not subscribed.
> 
> Yes thanks.  Likewise for our Fedora haskell-devel list though
> probably it is currently a bit too noisy with bugzilla mail traffic.
> 
> Actually I was wondering if it would make sense for us to
> setup a Haskell distro mailing-list for people involved/interesting
> in packaging haskell for Linux to communicate about issues?
> Perhaps even packaging for all platforms?  What do you think?

Not sure if we need it, and I’d rather wait until we know that for sure
before creating yet another list. I subscribed to haskell-devel, though,
and will skim over it from time to time.

> Thanks for reaching out and also I would like to say thanks and my respect
> for the great work Debian has done and continues to do on Haskell packaging.

Thanks for the kudos. We are also happy to have good competition :-)

Greetings,
Joachim

-- 
Joachim "nomeata" Breitner
Debian Developer
  nomeata@debian.org | ICQ# 74513189 | GPG-Keyid: 4743206C
  JID: nomeata@joachim-breitner.de | http://people.debian.org/~nomeata

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: