gcc 4.2.3 : make: *** [bootstrap] Error 2

gcc 4.2.3 : make: *** [bootstrap] Error 2

[Dennis Clarke]

This error occurs well into the stage 2 of the bootstrap process.

Here are the specifics :

configure line used was

../gcc-4.2.3/configure --with-as=/home/dclarke/local/bin/as
--with-ld=/home/dclarke/local/bin/ld --enable-threads=posix --disable-nls
--prefix=/home/dclarke/local --with-local-prefix=/home/dclarke/local
--enable-shared --enable-languages=c,c++,objc,fortran
--with-gmp=/home/dclarke/local --with-mpfr=/home/dclarke/local
--enable-bootstrap

GNU Binutils 2.18
GNU Make 3.81
flex 2.5.35
autoconf (GNU Autoconf) 2.61
automake (GNU automake) 1.10.1

Both GMP and MPFR are built and pass all tests and the libraries are created
fine :

gmp-4.2.2
mpfr-2.3.1

dclarke@vesta:~/build/first_pass/gcc/gcc-4.2.3-build$ ls -lap
$HOME/local/lib | grep -E "gmp|mpfr"
-rw-r--r--  1 dclarke csw  592016 Mar  9 17:48 libgmp.a
-rwxr-xr-x  1 dclarke csw     794 Mar  9 17:48 libgmp.la
lrwxrwxrwx  1 dclarke csw      15 Mar  9 17:48 libgmp.so -> libgmp.so.3.4.2
lrwxrwxrwx  1 dclarke csw      15 Mar  9 17:48 libgmp.so.3 -> libgmp.so.3.4.2
-rwxr-xr-x  1 dclarke csw  317869 Mar  9 17:48 libgmp.so.3.4.2
-rw-r--r--  1 dclarke csw 1812424 Mar  9 22:06 libmpfr.a
-rwxr-xr-x  1 dclarke csw     985 Mar  9 22:06 libmpfr.la
lrwxrwxrwx  1 dclarke csw      16 Mar  9 22:06 libmpfr.so -> libmpfr.so.1.1.1
lrwxrwxrwx  1 dclarke csw      16 Mar  9 22:06 libmpfr.so.1 -> libmpfr.so.1.1.1
-rwxr-xr-x  1 dclarke csw  985371 Mar  9 22:06 libmpfr.so.1.1.1
dclarke@vesta:~/build/first_pass/gcc/gcc-4.2.3-build$



existing GCC was :

dclarke@vesta:~/build/first_pass/gcc/gcc-4.2.3-build$ gcc -v
Using built-in specs.
Target: powerpc-linux-gnu
Configured with: ../src/configure -v
--enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr
--enable-shared --with-system-zlib --libexecdir=/usr/lib
--without-included-gettext --enable-threads=posix --enable-nls
--program-suffix=-4.1 --enable-__cxa_atexit --enable-clocale=gnu
--enable-libstdcxx-debug --enable-mpfr --disable-softfloat
--enable-targets=powerpc-linux,powerpc64-linux --with-cpu=default32
--enable-checking=release powerpc-linux-gnu
Thread model: posix
gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)

Stage 1 of the bootstrap process resulted in this :

dclarke@vesta:~/build/first_pass/gcc/gcc-4.2.3-build$
/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/gcc/xgcc -v
Using built-in specs.
Target: powerpc-unknown-linux-gnu
Configured with: ../gcc-4.2.3/configure --with-as=/home/dclarke/local/bin/as
--with-ld=/home/dclarke/local/bin/ld --enable-threads=posix --disable-nls
--prefix=/home/dclarke/local --with-local-prefix=/home/dclarke/local
--enable-shared --enable-languages=c,c++,objc,fortran
--with-gmp=/home/dclarke/local --with-mpfr=/home/dclarke/local
--enable-bootstrap
Thread model: posix
gcc version 4.2.3


The ERROR in stage 2 was thus :


checking for powerpc-unknown-linux-gnu-gfortran...
/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/./gcc/gfortran
-B/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/./gcc/
-B/home/dclarke/local/powerpc-unknown-linux-gnu/bin/
-B/home/dclarke/local/powerpc-unknown-linux-gnu/lib/ -isystem
/home/dclarke/local/powerpc-unknown-linux-gnu/include -isystem
/home/dclarke/local/powerpc-unknown-linux-gnu/sys-include
checking whether we are using the GNU Fortran compiler... no
checking whether
/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/./gcc/gfortran
-B/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/./gcc/
-B/home/dclarke/local/powerpc-unknown-linux-gnu/bin/
-B/home/dclarke/local/powerpc-unknown-linux-gnu/lib/ -isystem
/home/dclarke/local/powerpc-unknown-linux-gnu/include -isystem
/home/dclarke/local/powerpc-unknown-linux-gnu/sys-include accepts -g... no
checking whether the GNU Fortran compiler is working... no
configure: error: GNU Fortran is not working; the most common reason for
that is that you might have linked it to shared GMP and/or MPFR libraries,
and not set LD_LIBRARY_PATH accordingly. If you suspect any other reason,
please report a bug in http://gcc.gnu.org/bugzilla, attaching
/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/powerpc-unknown-linux-gnu/libgfortran/config.log
make[1]: *** [configure-target-libgfortran] Error 1
make[1]: Leaving directory `/home/dclarke/build/first_pass/gcc/gcc-4.2.3-build'
make: *** [bootstrap] Error 2
dclarke@vesta:~/build/first_pass/gcc/gcc-4.2.3-build$


This has happened repeatedly now .. twice actually. Not too sure what to do
to get out of this little bind.

Should I just set LD_LIBRARY_PATH=$HOME/local/lib  ??

Any thoughts ?

Dennis Clarke

Re: gcc 4.2.3 : make: *** [bootstrap] Error 2

[Ian Lance Taylor]

"Dennis Clarke" <dclarke@blastwave.org> writes:

> Both GMP and MPFR are built and pass all tests and the libraries are created
> fine :
>
> gmp-4.2.2
> mpfr-2.3.1
>
> dclarke@vesta:~/build/first_pass/gcc/gcc-4.2.3-build$ ls -lap
> $HOME/local/lib | grep -E "gmp|mpfr"
> -rw-r--r--  1 dclarke csw  592016 Mar  9 17:48 libgmp.a
> -rwxr-xr-x  1 dclarke csw     794 Mar  9 17:48 libgmp.la
> lrwxrwxrwx  1 dclarke csw      15 Mar  9 17:48 libgmp.so -> libgmp.so.3.4.2
> lrwxrwxrwx  1 dclarke csw      15 Mar  9 17:48 libgmp.so.3 -> libgmp.so.3.4.2
> -rwxr-xr-x  1 dclarke csw  317869 Mar  9 17:48 libgmp.so.3.4.2
> -rw-r--r--  1 dclarke csw 1812424 Mar  9 22:06 libmpfr.a
> -rwxr-xr-x  1 dclarke csw     985 Mar  9 22:06 libmpfr.la
> lrwxrwxrwx  1 dclarke csw      16 Mar  9 22:06 libmpfr.so -> libmpfr.so.1.1.1
> lrwxrwxrwx  1 dclarke csw      16 Mar  9 22:06 libmpfr.so.1 -> libmpfr.so.1.1.1
> -rwxr-xr-x  1 dclarke csw  985371 Mar  9 22:06 libmpfr.so.1.1.1


You will do better to use --disable-shared when you configure gmp and
mpfr.  Using the shared libraries means that you need to ensure that
gcc can find them at runtime, and that typically requires setting
LD_LIBRARY_PATH.


> /home/dclarke/local/powerpc-unknown-linux-gnu/sys-include accepts -g... no
> checking whether the GNU Fortran compiler is working... no
> configure: error: GNU Fortran is not working; the most common reason for
> that is that you might have linked it to shared GMP and/or MPFR libraries,
> and not set LD_LIBRARY_PATH accordingly. If you suspect any other reason,
> please report a bug in http://gcc.gnu.org/bugzilla, attaching
> /home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/powerpc-unknown-linux-gnu/libgfortran/config.log

There you go: set LD_LIBRARY_PATH to work around this, but even better
is to use static gmp and mpfr libraries.


> Should I just set LD_LIBRARY_PATH=$HOME/local/lib  ??

That will probably work too.

Ian

Re: gcc 4.2.3 : make: *** [bootstrap] Error 2

[Dennis Clarke]

> "Dennis Clarke" <dclarke@blastwave.org> writes:
>
>> Both GMP and MPFR are built and pass all tests and the libraries are
>> created fine :
>>
>> gmp-4.2.2
>> mpfr-2.3.1
>>
>
> You will do better to use --disable-shared when you configure gmp and
> mpfr.  Using the shared libraries means that you need to ensure that
> gcc can find them at runtime, and that typically requires setting
> LD_LIBRARY_PATH.

I'll swing around for a second pass at gmp and mpfr and rebuild them again.
I don't recall needing to do this in the past but .. anything is possible I
guess.

>> /home/dclarke/local/powerpc-unknown-linux-gnu/sys-include accepts -g... no
>> checking whether the GNU Fortran compiler is working... no
>> configure: error: GNU Fortran is not working; the most common reason for
>> that is that you might have linked it to shared GMP and/or MPFR libraries,
>> and not set LD_LIBRARY_PATH accordingly. If you suspect any other reason,
>> please report a bug in http://gcc.gnu.org/bugzilla, attaching
>> /home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/powerpc-unknown-linux-gnu/libgfortran/config.log
>
> There you go: set LD_LIBRARY_PATH to work around this, but even better
> is to use static gmp and mpfr libraries.
>
>
>> Should I just set LD_LIBRARY_PATH=$HOME/local/lib  ??
>
> That will probably work too.

I started a fresh bootstrap build about twelve hours ago with
LD_LIBRARY_PATH set and with some luck the process will continue to
completion and then onwards towards the testsuites.

Thank you and I'll keep plugging away at this.

Dennis Clarke

Re: gcc 4.2.3 : make: *** [bootstrap] Error 2

[Dennis Clarke]

> "Dennis Clarke" <dclarke@blastwave.org> writes:
>

>> /home/dclarke/local/powerpc-unknown-linux-gnu/sys-include accepts -g... no
>> checking whether the GNU Fortran compiler is working... no
>> configure: error: GNU Fortran is not working; the most common reason for
>> that is that you might have linked it to shared GMP and/or MPFR libraries,
>> and not set LD_LIBRARY_PATH accordingly. If you suspect any other reason,
>> please report a bug in http://gcc.gnu.org/bugzilla, attaching
>> /home/dclarke/build/first_pass/gcc/gcc-4.2.3-build/powerpc-unknown-linux-gnu/libgfortran/config.log
>
> There you go: set LD_LIBRARY_PATH to work around this, but even better
> is to use static gmp and mpfr libraries.
>
>
>> Should I just set LD_LIBRARY_PATH=$HOME/local/lib  ??
>
> That will probably work too.
>

nope ... my third ( or fourth .. I lost count ) attempt failed also, despite
having LD_LIBRARY_PATH set as above.

Same error.

The other idea on the table is :

> You will do better to use --disable-shared when you configure gmp and
> mpfr.  Using the shared libraries means that you need to ensure that
> gcc can find them at runtime, and that typically requires setting
> LD_LIBRARY_PATH.


So now I'll go back and rebuild gmp and mpfr with --disable-shared and then
try GCC 4.2.3 again.  That should take a day or so.

Stay tuned for futher fun with embedded PowerPC systems :-)

Dennis Clarke

Re: gcc 4.2.3 : make: *** [bootstrap] Error 2

[Dennis Clarke]

> "Dennis Clarke" <dclarke@blastwave.org> writes:
>
>> Both GMP and MPFR are built and pass all tests and the libraries are
>> created fine :
>>
>> gmp-4.2.2
>> mpfr-2.3.1

   woo hoo !

 Just a followup.

 I went back and rebuilt and tested gmp and mpfr as you suggested and
then started over with the bootstrap of gcc 4.2.3 and guess what? It went
all the way through the bootstrap process !

The testsuite is running and thus far I see :

                === gcc Summary ===

# of expected passes            42664
# of unexpected failures        6
# of expected failures          115
# of unresolved testcases       1
# of untested testcases         28
# of unsupported tests          402

that looks pretty darn good :-)

So I just wanted to come back and sya thank you. Something that is often
missing on maillists is someone coming back and saying hey, you helped me
out and it looks to be working fine.

Thank you !

Dennis Clarke