Hi,
Am Freitag, den 02.09.2011, 18:19 +1000 schrieb Erik de Castro Lopo:
> Joachim Breitner wrote:
>
> > In general it seems that the Haskell support for non-x86-architectures
> > is getting worse and worse :-(
>
> I know ghci is broken on PowerPC and is not yet working on Arm (GHC
> on Arm is very recent), but what other arches have broken a ghci?
>
> I've been working on and off on the PowerPC ghci problem but it would
> be interesting to know which other arches are broken and if they are
> broken in the same way.
well, I don’t know how broken it is, but it is disabled on a lot of
architectures:
# Whether to include GHCi in the compiler. Depends on whether the RTS linker
# has support for this OS/ARCH combination.
OsSupportsGHCi=$(strip $(patsubst $(HostOS_CPP), YES, $(findstring $(HostOS_CPP), mingw32 cygwin32 linux solaris2 freebsd dragonfly netbsd openbsd darwin)))
ArchSupportsGHCi=$(strip $(patsubst $(HostArch_CPP), YES, $(findstring $(HostArch_CPP), i386 x86_64 powerpc sparc sparc64)))
(from mk/config.mk.in).
Maybe ghci on armel is a low-hanging fruit, specially if this arch
selection was made before http://hackage.haskell.org/trac/ghc/ticket/631
was fixed (see last comment there).
That makes me also wonder if we can adjust the code in mk/config.mk for
kfreebsd-* and submit it upstream, so that we can remove the special
casing in debian/rules:
ifneq (,$(findstring $(DEB_HOST_ARCH), kfreebsd-i386 kfreebsd-amd64))
echo "GhcWithInterpreter=YES" >> mk/build.mk
endif
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