Build fails again
naur at post11.tele.dk
Fri Jun 22 10:46:38 EDT 2007
I am not sure which patches you mean. The build reported on were missing only
these patches in the ghc HEAD relative to "now" (20070622 at 1642 DK time):
Fri Jun 22 13:16:34 CEST 2007 Simon Marlow <simonmar at microsoft.com>
* FIX BUILD (on Windows): Cabal must invoke compiler/stage1/ghc-inplace
not compiler/ghc-inplace, which is a symlink.
Thu Jun 21 16:45:19 CEST 2007 Simon Marlow <simonmar at microsoft.com>
UNDO: FIX BUILD on Windows: horrible hack to work around make(?) bug
Fri Jun 22 15:12:11 CEST 2007 Simon Marlow <simonmar at microsoft.com>
* update with new libraries
And no patches were missing from the libraries.
I'll pull the missing patches from HEAD and try again.
On Friday 22 June 2007 15:32, Simon Marlow wrote:
> I'm pretty sure I fixed this one - do you definitely have the patches I made
> hsc2hs today?
> Thorkil Naur wrote:
> > A PPC Mac OS X build today ended like this:
> > Registering html-1.0.1...
> > rm -f -f stamp/configure.library.*.network network/unbuildable
> > ( cd network && setup/Setup configure \
> > \
> > --prefix='$topdir' \
> > --datadir='$prefix/share' \
> > --libsubdir='$compiler/lib/$pkgid' \
> > --with-compiler=../../compiler/ghc-inplace \
> > --with-hc-pkg=../../utils/ghc-pkg/ghc-pkg-inplace \
> > --with-hsc2hs=../../utils/hsc2hs/hsc2hs-inplace \
> > --with-ld=/usr/bin/ld \
> > --datasubdir=ghc \
> > --haddock-args="--use-contents=../index.html \
> > --use-index=../doc-index.html" \
> > --configure-option='--with-readline-libraries=/opt/local/lib'
> > \
> > --configure-option=--with-cc=gcc ) \
> > && touch stamp/configure.library.build.network || touch
> > network/unbuildable
> > configure: Reading installed packages...
> > Configuring network-2.0.1...
> > configure: Dependency base-any: using base-2.1
> > ...
> > configure: No greencard found
> > rm -f network/GNUmakefile
> > cp Makefile.local network
> > ifBuildable/ifBuildable network setup/Setup makefile -f GNUmakefile
> > Socket.hsc: In function 'main':
> > Socket.hsc:431: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:432: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:433: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:434: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:438: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:439: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:440: error: 'struct in6_addr' has no member named 's6_addr32'
> > Socket.hsc:441: error: 'struct in6_addr' has no member named 's6_addr32'
> > compiling dist/build/Network/Socket_hsc_make.c failed
> > command was: gcc -c -D__GLASGOW_HASKELL__=607
> > -Iinclude -DCALLCONV=ccall dist/build/Network/Socket_hsc_make.c -o
> > dist/build/Network/Socket_hsc_make.o
> > Reading parameters
> > Preprocessing library network-2.0.1...
> > make: *** [network/GNUmakefile] Error 1
> > make: *** [stage1] Error 2
> > Details at
> > Best regards
> > Thorkil
> > On Friday 22 June 2007 11:23, Simon Peyton-Jones wrote:
> >> This build stuff *will* settle down. Please keep reporting problems and
> > suggesting improvments.
> >> Simon's fix to the in-place stuff yesterday was a big step forward.
> > Previously we were generating both a *shell* script (on unix) and *.bat*
> > script (on windows). But it turns out that a MSYS shell can't invoke
> > script, and a cmd shell can't invoke a shell script, so it was a nightmare
> > make the MSYS build system work. Simon has a arranged to build an
> > instead, which always works, on any platform, invoked from any shell.
> > you Simon!
> >> Simon
> >> | -----Original Message-----
> >> | From: cvs-ghc-bounces at haskell.org [mailto:cvs-ghc-bounces at haskell.org]
> > Behalf Of Simon Marlow
> >> | Sent: 22 June 2007 09:42
> >> | To: Thorkil Naur
> >> | Cc: cvs-ghc at haskell.org; Manuel M T Chakravarty
> >> | Subject: Re: Build fails again
> >> |
> >> | Both of these problems are now fixed. I'll be running some more builds
> > today to
> >> | make sure the tree is buildable again.
More information about the Cvs-ghc