Re: CT: Any obstacles for using gcc 4.0 snapshots?

[Dan Kegel <dank at kegel dot com>]

Dan Kegel wrote:
> Martin Egholm Nielsen wrote:
>
> > > > Do you reckon I could make a gcc-4.0-20050305-glibc-2.2.5.dat for 
> > > > PPC405 without any complications?
> > >
> > >
> > > Heh. Try it and see.  Send a patch with the download fix and
> > > anything else you find.
> >
> >
> > I applied the patch that you submitted, and got to fetch and started 
> > compiling.
> > However, it died compiling glibc-2.2.5 with the following fault:
> >
> > === 8< 8< 8< ===
> > checking for powerpc-405-linux-gnu-gcc... powerpc-405-linux-gnu-gcc
> > checking version of powerpc-405-linux-gnu-gcc ... 4.0.0, bad
>
>
> I whipped up a little patch for that and am trying it now.

Next failure was:

gconv_open.c: In function '__gconv_open':
gconv_open.c:166: error: invalid lvalue in assignment
gconv_open.c:257: warning: pointer targets in assignment differ in signedness
make[2]: *** [/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/build-glibc/iconv/gconv_open.o] Error 1
make[2]: Leaving directory `/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/glibc-2.2.5/iconv'
make[1]: *** [iconv/subdir_lib] Error 2
make[1]: Leaving directory `/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/glibc-2.2.5'
make: *** [all] Error 2

which is the same problem fixed by
http://kegel.com/crosstool/crosstool-0.28-rc37/patches/glibc-2.3.2/glibc-2.3.2-allow-gcc-3.5-gconv.patch
So I rediffed that patch against glibc-2.2.5
and continued.

The next problem was

msort.c: In function 'msort_with_tmp':
msort.c:59: error: invalid lvalue in increment
msort.c:59: error: invalid lvalue in increment
msort.c:64: error: invalid lvalue in increment
msort.c:64: error: invalid lvalue in increment
make[2]: *** [/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/build-glibc/stdlib/msort.o] Error 1
make[2]: Leaving directory `/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/glibc-2.2.5/stdlib'
make[1]: *** [stdlib/subdir_lib] Error 2
make[1]: Leaving directory `/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/glibc-2.2.5'
make: *** [all] Error 2

the same problem solved by
http://kegel.com/crosstool/crosstool-0.28-rc37/patches/glibc-2.3.2/glibc-2.3.2-allow-gcc-3.5-msort.patch
So I rediffed that patch against glibc-2.2.5
and continued.

The next problem was
In file included from strcoll.c:101:
../locale/weight.h: In function 'strcoll':
../locale/weight.h:23: error: invalid storage class for function 'findidx'
make[2]: *** [/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/build-glibc/string/strcoll.o] Error 1
make[2]: Leaving directory `/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/glibc-2.2.5/string'
make[1]: *** [string/subdir_lib] Error 2
make[1]: Leaving directory `/home/dank/queue/jobdir.fast2/crosstool-dev/build/i686-unknown-linux-gnu/gcc-4.0-20050305-glibc-2.2.5/glibc-2.2.5'
make: *** [all] Error 2

I don't have a patch for this just yet, but it was discussed here
http://sources.redhat.com/ml/libc-hacker/2004-09/msg00015.html
and the fix checked in to cvs seems to be
http://sources.redhat.com/ml/glibc-cvs/2004-q3/msg01071.html

I'll probably get this all sorted out soon, but it does make
one think "gee, maybe using glibc-2.3.4 is a good idea
if I'm going to use gcc-4.0".
- Dan

--
Trying to get a job as a c++ developer?  See http://kegel.com/academy/getting-hired.html



------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com