On Fri, Aug 25, 2006 at 05:02:02PM +0200, Michel Dänzer <[EMAIL PROTECTED]>
wrote:
> On Fri, 2006-08-25 at 13:42 +0200, Michel Dänzer wrote:
> > Looking at a diff between build logs, I noticed that 1.8.0.5-1 built
> > xptc{stubs,invoke}_unsupported.cpp instead of
> > xptc{stubs,invoke}_ppc_linux.cpp. The cause is the following hunk in
> > 99_configure.dpatch:
> >
> >
> > @@ -5042,10 +4911,17 @@
> > darwin*) OS_ARCH=Darwin OS_TARGET=Darwin ;;
> > esac
> > else
> > - OS_TARGET=`uname -s`
> > - OS_ARCH=`uname -s | sed -e 's|/|_|g'`
> > + OS_TARGET="${host_os}"
> > + OS_ARCH=`echo $host_os | sed -e 's|/|_|g'`
> > OS_RELEASE=`uname -r`
> > - OS_TEST=`uname -m`
> > + OS_TEST="${host_cpu}"
> > + case "${host_os}" in
> > + linux*) OS_ARCH=Linux ;;
> > + solaris*) OS_ARCH=SunOS OS_RELEASE=5 ;;
> > + mingw*) OS_ARCH=WINNT ;;
> > + wince*) OS_ARCH=WINCE ;;
> > + darwin*) OS_ARCH=Darwin OS_TARGET=Darwin ;;
> > + esac
> > fi
> > _COMPILER_PREFIX=
> >
> >
> >
> > On powerpc, ${host_cpu} is 'powerpc', whereas the output of uname -m is
> > 'ppc' (as expected by the xulrunner code). I'm rebuilding 1.8.0.5-2 now
> > to verify that reversing this fixes the problem.
Thanks a lot for the analysis ! I'll try to upload a -3 version soon.
Mike
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
