gcc build fails "cannot compute suffix of object files" - Solaris

This is a discussion on gcc build fails "cannot compute suffix of object files" - Solaris ; I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express Community Edition Build 89. drkirkby@kingfisher:[~/build] $ cat /etc/release Solaris Express Community Edition snv_89 X86 Copyright 2008 Sun Microsystems, Inc. All Rights Reserved. Use is subject to license terms. ...

+ Reply to Thread
Page 1 of 3 1 2 3 LastLast
Results 1 to 20 of 57

Thread: gcc build fails "cannot compute suffix of object files"

  1. gcc build fails "cannot compute suffix of object files"

    I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    Community Edition Build 89.



    drkirkby@kingfisher:[~/build] $ cat /etc/release
    Solaris Express Community Edition snv_89 X86
    Copyright 2008 Sun Microsystems, Inc. All Rights Reserved.
    Use is subject to license terms.
    Assembled 06 May 2008


    After configuring with these options (plenty of others tried)

    $ ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/as
    --with-ld=/usr/ccs/bin/ld --enable-shared
    --enable-languages=c,c++,fortran LD_LIBRARY_PATH=/usr/local/lib
    --with-gmp=/usr/local --with-mpfr=/usr/local
    --with-mpfr-lib=/usr/local/lib --with-mpfr-include=/usr/local/include
    --with-gmp-include=/usr/local/include --with-gmp-lib=/usr/local/lib



    gcc spends 15 minutes or so compiling, then bombs out with:

    checking for i386-pc-solaris2.11-lipo... lipo
    checking for i386-pc-solaris2.11-nm... /export/home/drkirkby/build/./gcc/nm
    checking for i386-pc-solaris2.11-ranlib... ranlib
    checking for i386-pc-solaris2.11-strip... strip
    checking whether ln -s works... yes
    checking for i386-pc-solaris2.11-gcc...
    /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include
    checking for suffix of object files... configure: error: cannot compute
    suffix of object files: cannot compile
    See `config.log' for more details.
    gmake[2]: *** [configure-stage1-target-libgcc] Error 1
    gmake[2]: Leaving directory `/export/home/drkirkby/build'
    gmake[1]: *** [stage1-bubble] Error 2
    gmake[1]: Leaving directory `/ex


    A search on the web for the "configure: error: cannot compute suffix of
    object files: cannot compile" suggests setting LD_LIBRARY_PATH, as
    above. That does not work. I've also tried

    set LD_LIBRARY_PATH=/usr/local/lib
    export LD_LIBRARY_PATH

    but that does not work either.

    The log file:

    ../i386-pc-solaris2.11/libgcc/config.log
    (see end of message)

    shows the linker can't seem to find the library libmpfr.so.1. But that
    does exists. I've tried every option to configure that might tell the
    linker where to find this library:

    --with-mpfr=/usr/local
    --with-mpfr-lib=/usr/local/lib
    --with-mpfr-include=/usr/local/include

    but it does not work.

    In contrast, I can easily use that library if I compile with cc or gcc.

    Neither of these present any error message.
    drkirkby@kingfisher:[~] $ cc -L/usr/local/lib -lmpfr f.c
    drkirkby@kingfisher:[~] $ gcc -L/usr/local/lib -lmpfr f.c

    so the library is ok.

    Here's the log file. Any idea why the building of gcc 4.3.1 is failing?


    Target: i386-pc-solaris2.11
    Configured with: ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/as
    --with-ld=/usr/ccs/bin/ld --enable-shared
    --enable-languages=c,c++,fortran LD_LIBRARY_PATH=/usr/local/lib
    --with-gmp=/usr/local --with-mpfr=/usr/local
    --with-mpfr-lib=/usr/local/lib --with-mpfr-include=/usr/local/include
    --with-gmp-include=/usr/local/include --with-gmp-lib=/usr/local/lib
    Thread model: posix
    gcc version 4.3.1 (GCC)
    configure:2374: $? = 0
    configure:2376: /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -V &5
    xgcc: '-V' must come at the start of the command line
    configure:2379: $? = 1
    configure:2398: /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -o conftest -g
    -fkeep-inline-functions conftest.c >&5
    ld.so.1: cc1: fatal: libmpfr.so.1: open failed: No such file or directory
    xgcc: Internal error: Killed (program cc1)
    Please submit a full bug report.
    See for instructions.
    configure:2401: $? = 1
    configure:2567: checking for suffix of object files
    configure:2588: /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -c -g -fkeep-inline-functions
    conftest.c >&5
    ld.so.1: cc1: fatal: libmpfr.so.1: open failed: No such file or directory
    xgcc: Internal error: Killed (program cc1)
    Please submit a full bug report.
    See for instructions.
    configure:2591: $? = 1
    configure: failed program was:


  2. Re: gcc build fails "cannot compute suffix of object files"

    Dave wrote:
    > I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    > Community Edition Build 89.


    Looking at my log, I overlooked to mention this line - see full lot
    posted previously.

    xgcc: Internal error: Killed (program cc1)


    Any ideas how to solve it?

  3. Re: gcc build fails "cannot compute suffix of object files"

    Dave schrieb:
    > I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    > Community Edition Build 89.
    >
    >
    >
    > drkirkby@kingfisher:[~/build] $ cat /etc/release
    > Solaris Express Community Edition snv_89 X86
    > Copyright 2008 Sun Microsystems, Inc. All Rights Reserved.
    > Use is subject to license terms.
    > Assembled 06 May 2008
    >
    >
    > After configuring with these options (plenty of others tried)
    >
    > $ ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/as
    > --with-ld=/usr/ccs/bin/ld --enable-shared
    > --enable-languages=c,c++,fortran LD_LIBRARY_PATH=/usr/local/lib
    > --with-gmp=/usr/local --with-mpfr=/usr/local
    > --with-mpfr-lib=/usr/local/lib --with-mpfr-include=/usr/local/include
    > --with-gmp-include=/usr/local/include --with-gmp-lib=/usr/local/lib
    >
    >
    >
    > gcc spends 15 minutes or so compiling, then bombs out with:
    >
    > checking for i386-pc-solaris2.11-lipo... lipo
    > checking for i386-pc-solaris2.11-nm... /export/home/drkirkby/build/./gcc/nm
    > checking for i386-pc-solaris2.11-ranlib... ranlib
    > checking for i386-pc-solaris2.11-strip... strip
    > checking whether ln -s works... yes
    > checking for i386-pc-solaris2.11-gcc...
    > /export/home/drkirkby/build/./gcc/xgcc
    > -B/export/home/drkirkby/build/./gcc/
    > -B/usr/local/i386-pc-solaris2.11/bin/
    > -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    > /usr/local/i386-pc-solaris2.11/include -isystem
    > /usr/local/i386-pc-solaris2.11/sys-include
    > checking for suffix of object files... configure: error: cannot compute
    > suffix of object files: cannot compile
    > See `config.log' for more details.
    > gmake[2]: *** [configure-stage1-target-libgcc] Error 1
    > gmake[2]: Leaving directory `/export/home/drkirkby/build'
    > gmake[1]: *** [stage1-bubble] Error 2
    > gmake[1]: Leaving directory `/ex
    >
    >
    > A search on the web for the "configure: error: cannot compute suffix of
    > object files: cannot compile" suggests setting LD_LIBRARY_PATH, as
    > above. That does not work. I've also tried
    >
    > set LD_LIBRARY_PATH=/usr/local/lib
    > export LD_LIBRARY_PATH
    >
    > but that does not work either.
    >
    > The log file:
    >
    > ./i386-pc-solaris2.11/libgcc/config.log
    > (see end of message)
    >
    > shows the linker can't seem to find the library libmpfr.so.1. But that
    > does exists. I've tried every option to configure that might tell the
    > linker where to find this library:
    >
    > --with-mpfr=/usr/local
    > --with-mpfr-lib=/usr/local/lib
    > --with-mpfr-include=/usr/local/include
    >
    > but it does not work.
    >
    > In contrast, I can easily use that library if I compile with cc or gcc.
    >
    > Neither of these present any error message.
    > drkirkby@kingfisher:[~] $ cc -L/usr/local/lib -lmpfr f.c
    > drkirkby@kingfisher:[~] $ gcc -L/usr/local/lib -lmpfr f.c
    >
    > so the library is ok.
    >
    > Here's the log file. Any idea why the building of gcc 4.3.1 is failing?
    >
    >
    > Target: i386-pc-solaris2.11
    > Configured with: ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/as
    > --with-ld=/usr/ccs/bin/ld --enable-shared
    > --enable-languages=c,c++,fortran LD_LIBRARY_PATH=/usr/local/lib
    > --with-gmp=/usr/local --with-mpfr=/usr/local
    > --with-mpfr-lib=/usr/local/lib --with-mpfr-include=/usr/local/include
    > --with-gmp-include=/usr/local/include --with-gmp-lib=/usr/local/lib
    > Thread model: posix
    > gcc version 4.3.1 (GCC)
    > configure:2374: $? = 0
    > configure:2376: /export/home/drkirkby/build/./gcc/xgcc
    > -B/export/home/drkirkby/build/./gcc/
    > -B/usr/local/i386-pc-solaris2.11/bin/
    > -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    > /usr/local/i386-pc-solaris2.11/include -isystem
    > /usr/local/i386-pc-solaris2.11/sys-include -V &5
    > xgcc: '-V' must come at the start of the command line
    > configure:2379: $? = 1
    > configure:2398: /export/home/drkirkby/build/./gcc/xgcc
    > -B/export/home/drkirkby/build/./gcc/
    > -B/usr/local/i386-pc-solaris2.11/bin/
    > -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    > /usr/local/i386-pc-solaris2.11/include -isystem
    > /usr/local/i386-pc-solaris2.11/sys-include -o conftest -g
    > -fkeep-inline-functions conftest.c >&5
    > ld.so.1: cc1: fatal: libmpfr.so.1: open failed: No such file or directory
    > xgcc: Internal error: Killed (program cc1)
    > Please submit a full bug report.
    > See for instructions.
    > configure:2401: $? = 1
    > configure:2567: checking for suffix of object files
    > configure:2588: /export/home/drkirkby/build/./gcc/xgcc
    > -B/export/home/drkirkby/build/./gcc/
    > -B/usr/local/i386-pc-solaris2.11/bin/
    > -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    > /usr/local/i386-pc-solaris2.11/include -isystem
    > /usr/local/i386-pc-solaris2.11/sys-include -c -g -fkeep-inline-functions
    > conftest.c >&5
    > ld.so.1: cc1: fatal: libmpfr.so.1: open failed: No such file or directory
    > xgcc: Internal error: Killed (program cc1)
    > Please submit a full bug report.
    > See for instructions.
    > configure:2591: $? = 1
    > configure: failed program was:
    >


    Builing gcc on Solaris is a bit tricky...

    IIRC, gcc 4.3.1 requires CONFIG_SHELL to be set to /bin/bash and LDFLAGS
    to be set to include libmpfr's libdir path in the RUNPATH variable i.e.
    'LDFLAGS="-Xlinker -R/usr/local/lib"'

    BTW: I was unable into tricking gcc's build procedure to include
    libmpfr's path into gcc's RUNPATH...

    If this doesn't work, I can try looking up at home, how I managed to
    build it...

    HTH,
    Thomas

  4. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 15:57:45 +0200, Thomas Maier-Komor wrote:

    > Builing gcc on Solaris is a bit tricky...


    Understatement of the day!

    > IIRC, gcc 4.3.1 requires CONFIG_SHELL to be set to /bin/bash and LDFLAGS
    > to be set to include libmpfr's libdir path in the RUNPATH variable i.e.
    > 'LDFLAGS="-Xlinker -R/usr/local/lib"'


    Or just have the builder use bash.

    LDFLAGS='-L/usr/local/lib -R/usr/local/lib' should work. It did for me
    when building gcc-4.1.2.

    Never set LD_LIBRARY_PATH, specially when compiling! Always use LDFLAGS
    with both -L/path and -R/path.

    > BTW: I was unable into tricking gcc's build procedure to include
    > libmpfr's path into gcc's RUNPATH...


    Probably better to set gcc prefix to the same directory where gmp and mpfr
    reside.

    Additionally, be certain to unset CFLAGS before running the configure
    script.


  5. Re: gcc build fails "cannot compute suffix of object files"

    Dave wrote:
    > A search on the web for the "configure: error: cannot compute suffix of
    > object files: cannot compile" suggests setting LD_LIBRARY_PATH, as
    > above. That does not work. I've also tried
    >
    > set LD_LIBRARY_PATH=/usr/local/lib
    > export LD_LIBRARY_PATH
    >
    > but that does not work either.
    >
    > The log file:
    >
    > ./i386-pc-solaris2.11/libgcc/config.log
    > (see end of message)
    >
    > shows the linker can't seem to find the library libmpfr.so.1. But that
    > does exists. I've tried every option to configure that might tell the


    Try LD_OPTIONS (see "man ld")

    LD_OPTIONS="-L/usr/local/lib -R/usr/local/lib"
    export LD_OPTIONS

    or for csh/tcsh:
    setenv LD_OPTIONS "-L/usr/local/lib -R/usr/local/lib"

  6. Re: gcc build fails "cannot compute suffix of object files"

    Dave Uhring schrieb:

    > LDFLAGS='-L/usr/local/lib -R/usr/local/lib' should work. It did for me
    > when building gcc-4.1.2.
    >


    -L/usr/local/lib should be automatically set by configure when passing
    --with-mpfr=/usr/local.

    -R/usr/local/lib will not work for gcc without -Xlinker.

  7. Re: gcc build fails "cannot compute suffix of object files"

    Thomas Maier-Komor wrote:
    >
    >
    > Builing gcc on Solaris is a bit tricky...


    I can see that. It seems to be black magic to ring the right configure
    options.

    > IIRC, gcc 4.3.1 requires CONFIG_SHELL to be set to /bin/bash and LDFLAGS
    > to be set to include libmpfr's libdir path in the RUNPATH variable i.e.
    > 'LDFLAGS="-Xlinker -R/usr/local/lib"'


    Tried both of them together, and it still fails.

    > BTW: I was unable into tricking gcc's build procedure to include
    > libmpfr's path into gcc's RUNPATH...
    >
    > If this doesn't work, I can try looking up at home, how I managed to
    > build it...


    I'd be intersted in the options you gave.

    I've tried to build with gcc 4.0.2 (from Blastwave) and gcc 3.4.1 from
    /usr/sfw/


    This is my last (of many) attempts

    ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/
    --with-ld=/usr/ccs/bin/ld -enable-languages=c,c++,fortran
    --with-gmp=/usr/local --with-mpfr=/usr/local LDFLAGS="-L/usr/local/lib
    -R/usr/local/lib" CC=/usr/sfw/bin/gcc CXX=/usr/sfw/bin/g++
    CONFIG_SHELL=/bin/bash


    I've also tried building with the Sun compiler, but get another sort of
    error there - I forget what that error was about.

    Also tried:

    $ LD_OPTIONS="-L/usr/local/lib -R/usr/local/lib"
    $ export LD_OPTIONS



    >
    > HTH,
    > Thomas


  8. Re: gcc build fails "cannot compute suffix of object files"

    Dave schrieb:
    > Thomas Maier-Komor wrote:
    >>
    >>
    >> Builing gcc on Solaris is a bit tricky...

    >
    > I can see that. It seems to be black magic to ring the right configure
    > options.
    >
    >> IIRC, gcc 4.3.1 requires CONFIG_SHELL to be set to /bin/bash and
    >> LDFLAGS to be set to include libmpfr's libdir path in the RUNPATH
    >> variable i.e. 'LDFLAGS="-Xlinker -R/usr/local/lib"'

    >
    > Tried both of them together, and it still fails.
    >
    >> BTW: I was unable into tricking gcc's build procedure to include
    >> libmpfr's path into gcc's RUNPATH...
    >>
    >> If this doesn't work, I can try looking up at home, how I managed to
    >> build it...

    >
    > I'd be intersted in the options you gave.
    >
    > I've tried to build with gcc 4.0.2 (from Blastwave) and gcc 3.4.1 from
    > /usr/sfw/
    >
    >
    > This is my last (of many) attempts
    >
    > ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/
    > --with-ld=/usr/ccs/bin/ld -enable-languages=c,c++,fortran
    > --with-gmp=/usr/local --with-mpfr=/usr/local LDFLAGS="-L/usr/local/lib
    > -R/usr/local/lib" CC=/usr/sfw/bin/gcc CXX=/usr/sfw/bin/g++
    > CONFIG_SHELL=/bin/bash
    >


    you omitted -Xlinker before -R...

    Forget about trying to compile gcc with Sun's cc. That simply does not
    work...

    - Thomas

  9. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 09:12:47 -0500, Dave Uhring wrote:
    > On Mon, 09 Jun 2008 15:57:45 +0200, Thomas Maier-Komor wrote:
    >> Builing gcc on Solaris is a bit tricky...

    > Understatement of the day!


    Well that may be the case for x86 but the pre-requisites for this gcc are
    somewhat of an ugly build if you want to have 32/64 bit support.

    Once I packed all those away in the right spots and made them shareable
    ONLY I [hate static libs] found the configure troublesome only in that I
    had to specify where libmpfr and libgmp were with the suggested flags. No
    other configure incantation worked (that usually do).

    >> IIRC, gcc 4.3.1 requires CONFIG_SHELL to be set to /bin/bash and
    >> LDFLAGS to be set to include libmpfr's libdir path in the RUNPATH
    >> variable i.e. 'LDFLAGS="-Xlinker -R/usr/local/lib"'


    Nope - I had to use --with switches

    > Or just have the builder use bash.


    configure scripts often assume this : <

    > LDFLAGS='-L/usr/local/lib -R/usr/local/lib' should work. It did for me
    > when building gcc-4.1.2.


    I didnt need it

    > Never set LD_LIBRARY_PATH, specially when compiling! Always use LDFLAGS
    > with both -L/path and -R/path.


    Actually its LD_RUN_PATH that helps the linker which is really what you
    want.

    >> BTW: I was unable into tricking gcc's build procedure to include
    >> libmpfr's path into gcc's RUNPATH...


    See above : >

    > Probably better to set gcc prefix to the same directory where gmp and
    > mpfr reside.


    Thats what i did as its likely Id never use it anyway except for gcc,
    I put the lot in /opt/gcc

    > Additionally, be certain to unset CFLAGS before running the configure
    > script.


    I didnt bother. No issues yet with a gmake bootstrap but theres a LONG way
    to go still.


  10. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 14:25:42 +0100, Dave wrote:

    > Dave wrote:
    >> I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    >> Community Edition Build 89.

    Me on b90 SPARC
    > Looking at my log, I overlooked to mention this line - see full lot
    > posted previously.
    > xgcc: Internal error: Killed (program cc1)
    > Any ideas how to solve it?


    Depends on what options are causing the croak. Usually its -O something
    where the optimizer fails miserably. This is TOO common IMO,
    but it could be anything. It should not happen but if it does
    you have to try building it by hand one .o at a time.

    Why O why does the Doc need this haywire of compiler : <




  11. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 13:18:04 -0700, AGT wrote:
    > On Mon, 09 Jun 2008 09:12:47 -0500, Dave Uhring wrote:


    >> Never set LD_LIBRARY_PATH, specially when compiling! Always use LDFLAGS
    >> with both -L/path and -R/path.

    >
    > Actually its LD_RUN_PATH that helps the linker which is really what you
    > want.


    ld(1)

    LD_RUN_PATH An alternative mechanism for specifying a
    runpath to the link-editor. See the -R
    option. If both LD_RUN_PATH and the -R
    option are specified, -R supersedes.

    >> Additionally, be certain to unset CFLAGS before running the configure
    >> script.

    >
    > I didnt bother. No issues yet with a gmake bootstrap but theres a LONG way
    > to go still.


    My usual CFLAGS for gcc-3.4.3 are '-O2 -pipe -m32'. That did cause
    problems late in the gcc-4 build.


  12. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 14:25:42 +0100, Dave wrote:

    > Dave wrote:
    >> I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    >> Community Edition Build 89.

    > Looking at my log, I overlooked to mention this line - see full lot
    > posted previously.
    > xgcc: Internal error: Killed (program cc1)
    > Any ideas how to solve it?


    You have to read the config.log. As suggested. I just got this "error" too
    when bootstrapping in libgcc's configure stage. According to the log
    libgmp and libmpfr cannot be found. Which is of course ridiculous as - up
    til now anyway - there has NOT been a problem finding the stuff. I had to
    start from scratch with MORE configure options, Im adding LDFLAGS.
    Eventually this piece of crap will compile ??



  13. Re: gcc build fails "cannot compute suffix of object files"

    AGT wrote:
    > On Mon, 09 Jun 2008 14:25:42 +0100, Dave wrote:
    >
    >> Dave wrote:
    >>> I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    >>> Community Edition Build 89.

    >> Looking at my log, I overlooked to mention this line - see full lot
    >> posted previously.
    >> xgcc: Internal error: Killed (program cc1)
    >> Any ideas how to solve it?

    >
    > You have to read the config.log. As suggested. I just got this "error" too
    > when bootstrapping in libgcc's configure stage. According to the log
    > libgmp and libmpfr cannot be found. Which is of course ridiculous as - up
    > til now anyway - there has NOT been a problem finding the stuff. I had to
    > start from scratch with MORE configure options, Im adding LDFLAGS.
    > Eventually this piece of crap will compile ??
    >
    >



    A look on the 'Successful builds' page for this release of gcc:

    http://gcc.gnu.org/gcc-4.3/buildstat.html

    shows only SPARC and then only Solaris 2.6 and 2.8. Nothing about
    Solaris on x86, or any modern version of Solaris. Clearly the GNU
    developers do not take portability too seriously.

    It's hard to believe they would release a version of gcc without testing
    it properly on at the very least Solaris 10 on x86 and SPARC.

    I suspect the problem in the usual one - people don't make the effort to
    check that code builds properly on other platforms.

    I think Sun could do a lot here though, by making Sun hardware available
    to software developers for free by remote access. If Sun did that, there
    might be a few more gcc developers who would take the trouble to test
    the code on Solaris.

    HP offer the TestDrive site

    http://www.testdrive.hp.com/

    where anyone can get an account on a range of HP machines to test
    software. I've used it myself several times.

    When I developed 'ATLC'

    http://atlc.sourceforge.net/

    I made sure it compiled on pretty much anything. Solaris, Linux, AIX,
    HP-UX, Tru64, IRIX, UNICOS, NetBSD, OpenBSD, FreeBSD etc etc. The
    biggest hassle was an old Cray, where sizeof(short) == 8.

    I'm trying to help with a port of the maths program Sage

    http://www.sagemath.org/

    to Solaris and have made a Sun Ultra 60 available to a couple of
    developers, so they can check their changes on Solaris. Hopefully there
    will be a Solaris port of Sage within days, or at the most weeks.




  14. Re: gcc build fails "cannot compute suffix of object files"

    Thomas Maier-Komor wrote:

    > Builing gcc on Solaris is a bit tricky...


    Since nobody has managed to suggest a solution, I've reported this as a
    bug at

    http://gcc.gnu.org/bugzilla/

    Bug 36481




  15. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 15:42:20 -0500, Dave Uhring wrote:

    > On Mon, 09 Jun 2008 13:18:04 -0700, AGT wrote:
    >> On Mon, 09 Jun 2008 09:12:47 -0500, Dave Uhring wrote:

    >
    >>> Never set LD_LIBRARY_PATH, specially when compiling! Always use LDFLAGS
    >>> with both -L/path and -R/path.

    >> Actually its LD_RUN_PATH that helps the linker which is really what you
    >> want.

    > ld(1)
    >
    > LD_RUN_PATH An alternative mechanism for specifying a
    > runpath to the link-editor. See the -R
    > option. If both LD_RUN_PATH and the -R
    > option are specified, -R supersedes.


    Yes - and unfortunately you may have to specify BOTH or "configure" may
    not behave correctly and you will get the error Doctor Dave and I got.
    LD_RUN_PATH normally works just fine, however during the bootstrap build
    of libgcc LD_RUN_PATH (which I set with "env") is not inherited (!)

    Anyway I guess my point is that you have to be prepared for all kinds of
    nonsense with this "configure" rubbish. But anyone working with GNU stuff
    on Solaris knows this already.

    Hours have passed and Im now at a c++ 32 bit shared build stage..
    So setting LDFLAGS turned the trick it seems. So far.

    >>> Additionally, be certain to unset CFLAGS before running the configure
    >>> script.

    >> I didnt bother. No issues yet with a gmake bootstrap but theres a LONG
    >> way to go still.

    > My usual CFLAGS for gcc-3.4.3 are '-O2 -pipe -m32'. That did cause
    > problems late in the gcc-4 build.


    Then you arent building a multilib version?
    Maybe on x86 thats still a problem


  16. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 18:25:38 -0700, AGT wrote:
    > On Mon, 09 Jun 2008 15:42:20 -0500, Dave Uhring wrote:


    > Anyway I guess my point is that you have to be prepared for all kinds of
    > nonsense with this "configure" rubbish. But anyone working with GNU stuff
    > on Solaris knows this already.


    Normally the GNU autoconf generated configure scripts are pretty well
    behaved and produce usable Makefiles and config.h. Version 4 of gcc is
    anything but well behaved.

    > Hours have passed and Im now at a c++ 32 bit shared build stage.. So
    > setting LDFLAGS turned the trick it seems. So far.


    You may yet find a problem in building libstdc++ when the build produces
    64 bit objects but expects to link them with 32 bit objects.

    >> My usual CFLAGS for gcc-3.4.3 are '-O2 -pipe -m32'. That did cause
    >> problems late in the gcc-4 build.

    >
    > Then you arent building a multilib version? Maybe on x86 thats still a
    > problem


    When trying to build gcc I unset CFLAGS. I build all apps and their
    dependent libraries in 32 bit mode only but the gcc build process produces
    both ELFCLASSES of objects. It just gets confused about what it is
    building.

    I did try building gcc-4 only as an exercise. The GNU compiler which
    ships with Solaris is quite adequate for building even the latest version
    of the qt library, KDE and everything else which I wanted available and
    which would not build with Studio12. qt will build with Studio12 but KDE
    pukes with Sun's compiler.


  17. Re: gcc build fails "cannot compute suffix of object files"

    Dave Uhring wrote:

    > I did try building gcc-4 only as an exercise. The GNU compiler which
    > ships with Solaris is quite adequate for building even the latest version
    > of the qt library, KDE and everything else which I wanted available and
    > which would not build with Studio12. qt will build with Studio12 but KDE
    > pukes with Sun's compiler.
    >



    According to the docs on the statistics package R,

    http://cran.r-project.org/doc/manuals/R-admin.pdf

    it fails some tests if built with gcc 4 on Solaris x86 and I needed to
    use 4.x as part of something larger. I found that to be true too - it
    passes on SPARC, but not on x86.

    Hence I thought a later version of gcc might be better, but it appears
    that perhaps this is not so.





  18. Re: gcc build fails "cannot compute suffix of object files"

    AGT wrote:
    > On Mon, 09 Jun 2008 15:42:20 -0500, Dave Uhring wrote:
    >
    >> On Mon, 09 Jun 2008 13:18:04 -0700, AGT wrote:
    >>> On Mon, 09 Jun 2008 09:12:47 -0500, Dave Uhring wrote:
    >>>> Never set LD_LIBRARY_PATH, specially when compiling! Always use LDFLAGS
    >>>> with both -L/path and -R/path.
    >>> Actually its LD_RUN_PATH that helps the linker which is really what you
    >>> want.

    >> ld(1)
    >>
    >> LD_RUN_PATH An alternative mechanism for specifying a
    >> runpath to the link-editor. See the -R
    >> option. If both LD_RUN_PATH and the -R
    >> option are specified, -R supersedes.

    >
    > Yes - and unfortunately you may have to specify BOTH or "configure" may
    > not behave correctly and you will get the error Doctor Dave and I got.
    > LD_RUN_PATH normally works just fine, however during the bootstrap build
    > of libgcc LD_RUN_PATH (which I set with "env") is not inherited (!)
    >
    > Anyway I guess my point is that you have to be prepared for all kinds of
    > nonsense with this "configure" rubbish. But anyone working with GNU stuff
    > on Solaris knows this already.
    >
    > Hours have passed and Im now at a c++ 32 bit shared build stage..
    > So setting LDFLAGS turned the trick it seems. So far.



    Since you seem to be getting further than me, as you are running for
    hours and mine is bombing out after only 5 minutes, what are use using?
    I've tried:

    $ mkdir build
    $ cd build
    $ LD_RUN_PATH=/usr/local/lib
    $ export LD_RUN_PATH
    $ LDFLAGS="-L/usr/local/lib -R/usr/local/lib"
    $ export LDFLAGS
    $ ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/as
    --with-ld=/usr/ccs/bin/ld --enable-shared
    --enable-languages=c,c++,fortran LD_LIBRARY_PATH=/usr/local/lib
    --with-gmp=/usr/local --with-mpfr=/usr/local
    --with-mpfr-lib=/usr/local/lib --with-mpfr-include=/usr/local/include
    --with-gmp-include=/usr/local/include --with-gmp-lib=/usr/local/lib
    LD_RUN_PATH=/usr/local/lib
    $ gmake bootstrap

    and *still* gcc 4.3.1 bombs out with:


    I dare not try to build this on my power hungry Blade 2000 - I think I
    will die of old age waiting for it to compiler if I don''t use my
    battery powered laptop!


  19. Re: gcc build fails "cannot compute suffix of object files"

    Dave wrote:

    >> Hours have passed and Im now at a c++ 32 bit shared build stage..
    >> So setting LDFLAGS turned the trick it seems. So far.

    >
    >
    > Since you seem to be getting further than me, as you are running for
    > hours and mine is bombing out after only 5 minutes, what are use using?
    > I've tried:
    >
    > $ mkdir build
    > $ cd build
    > $ LD_RUN_PATH=/usr/local/lib
    > $ export LD_RUN_PATH
    > $ LDFLAGS="-L/usr/local/lib -R/usr/local/lib"
    > $ export LDFLAGS
    > $ ../gcc-4.3.1/configure --with-as=/usr/ccs/bin/as
    > --with-ld=/usr/ccs/bin/ld --enable-shared
    > --enable-languages=c,c++,fortran LD_LIBRARY_PATH=/usr/local/lib
    > --with-gmp=/usr/local --with-mpfr=/usr/local
    > --with-mpfr-lib=/usr/local/lib --with-mpfr-include=/usr/local/include
    > --with-gmp-include=/usr/local/include --with-gmp-lib=/usr/local/lib
    > LD_RUN_PATH=/usr/local/lib
    > $ gmake bootstrap
    >
    > and *still* gcc 4.3.1 bombs out with:
    >
    >
    > I dare not try to build this on my power hungry Blade 2000 - I think I
    > will die of old age waiting for it to compiler if I don''t use my
    > battery powered laptop!
    >




    Oops,

    I was composing that email and intended sending it once it bombed out.
    BUT it got a bit further this time. I rekon it took 10 minutes to fail,
    not 5.

    It fails differently this time.




    02"
    "/var/tmp//ccPzaO8Z.s", line 8004 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB28", ".DOT-103"
    "/var/tmp//ccPzaO8Z.s", line 8027 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB55", ".DOT-104"
    "/var/tmp//ccPzaO8Z.s", line 8060 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB56", ".DOT-105"
    /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -g -fkeep-inline-functions
    -m64 -O2 -O2 -g -g -O2 -DIN_GCC -W -Wall -Wwrite-strings
    -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition
    -isystem ./include -fPIC -g -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2
    -D__GCC_FLOAT_NOT_NEEDED -I. -I. -I../../.././gcc
    -I../../../../gcc-4.3.1/libgcc -I../../../../gcc-4.3.1/libgcc/.
    -I../../../../gcc-4.3.1/libgcc/../gcc
    -I../../../../gcc-4.3.1/libgcc/../include -DHAVE_CC_TLS -o
    unwind-sjlj_s.o -MT unwind-sjlj_s.o -MD -MP -MF unwind-sjlj_s.dep
    -DSHARED -fexceptions -c ../../../../gcc-4.3.1/libgcc/../gcc/unwind-sjlj.c
    Assembler: unwind-sjlj.c
    "/var/tmp//cc18aa9Z.s", line 1283 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB2", ".DOT-17"
    "/var/tmp//cc18aa9Z.s", line 1292 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB3", ".DOT-18"
    "/var/tmp//cc18aa9Z.s", line 1311 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB5", ".DOT-19"
    "/var/tmp//cc18aa9Z.s", line 1320 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB6", ".DOT-20"
    "/var/tmp//cc18aa9Z.s", line 1329 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB7", ".DOT-21"
    "/var/tmp//cc18aa9Z.s", line 1338 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB8", ".DOT-22"
    "/var/tmp//cc18aa9Z.s", line 1347 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB16", ".DOT-23"
    "/var/tmp//cc18aa9Z.s", line 1356 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB17", ".DOT-24"
    "/var/tmp//cc18aa9Z.s", line 1365 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB18", ".DOT-25"
    "/var/tmp//cc18aa9Z.s", line 1374 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB12", ".DOT-26"
    "/var/tmp//cc18aa9Z.s", line 1399 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB11", ".DOT-27"
    "/var/tmp//cc18aa9Z.s", line 1422 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB10", ".DOT-28"
    "/var/tmp//cc18aa9Z.s", line 1445 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB9", ".DOT-29"
    "/var/tmp//cc18aa9Z.s", line 1468 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB4", ".DOT-30"
    "/var/tmp//cc18aa9Z.s", line 1493 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB15", ".DOT-31"
    "/var/tmp//cc18aa9Z.s", line 1516 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB13", ".DOT-32"
    "/var/tmp//cc18aa9Z.s", line 1539 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB14", ".DOT-33"
    /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -g -fkeep-inline-functions
    -m64 -O2 -O2 -g -g -O2 -DIN_GCC -W -Wall -Wwrite-strings
    -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition
    -isystem ./include -fPIC -g -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2
    -D__GCC_FLOAT_NOT_NEEDED -I. -I. -I../../.././gcc
    -I../../../../gcc-4.3.1/libgcc -I../../../../gcc-4.3.1/libgcc/.
    -I../../../../gcc-4.3.1/libgcc/../gcc
    -I../../../../gcc-4.3.1/libgcc/../include -DHAVE_CC_TLS -o
    gthr-gnat_s.o -MT gthr-gnat_s.o -MD -MP -MF gthr-gnat_s.dep -DSHARED
    -fexceptions -c ../../../../gcc-4.3.1/libgcc/../gcc/gthr-gnat.c
    Assembler: gthr-gnat.c
    "/var/tmp//ccKcay9Z.s", line 217 : Warning: Illegal subtraction
    - symbols from different sections: ".LFB2", ".DOT-6"
    "/var/tmp//ccKcay9Z.s", line 226 : Warning: Illegal subtraction
    - symbols from different sections: ".LFB3", ".DOT-7"
    "/var/tmp//ccKcay9Z.s", line 235 : Warning: Illegal subtraction
    - symbols from different sections: ".LFB4", ".DOT-8"
    "/var/tmp//ccKcay9Z.s", line 244 : Warning: Illegal subtraction
    - symbols from different sections: ".LFB5", ".DOT-9"
    "/var/tmp//ccKcay9Z.s", line 253 : Warning: Illegal subtraction
    - symbols from different sections: ".LFB6", ".DOT-10"
    "/var/tmp//ccKcay9Z.s", line 272 : Warning: Illegal subtraction -
    symbols from different sections: ".LFB7",
    ".DOT-11"/export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -g -fkeep-inline-functions
    -m64 -O2 -O2 -g -g -O2 -DIN_GCC -W -Wall -Wwrite-strings
    -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition
    -isystem ./include -fPIC -g -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2
    -D__GCC_FLOAT_NOT_NEEDED -I. -I. -I../../.././gcc
    -I../../../../gcc-4.3.1/libgcc -I../../../../gcc-4.3.1/libgcc/.
    -I../../../../gcc-4.3.1/libgcc/../gcc
    -I../../../../gcc-4.3.1/libgcc/../include -DHAVE_CC_TLS -o unwind-c_s.o
    -MT unwind-c_s.o -MD -MP -MF unwind-c_s.dep -DSHARED -fexceptions -c
    .../../../../gcc-4.3.1/libgcc/../gcc/unwind-c.c
    Assembler: unwind-c.c
    "/var/tmp//ccNdaW9Z.s", line 1845 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB6", ".DOT-2"
    "/var/tmp//ccNdaW9Z.s", line 1876 : Warning: Illegal subtraction -
    symbols from different sections: ".LFB8",
    ".DOT-3"/export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -g -fkeep-inline-functions
    -m64 -O2 -O2 -g -g -O2 -DIN_GCC -W -Wall -Wwrite-strings
    -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition
    -isystem ./include -fPIC -g -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2
    -D__GCC_FLOAT_NOT_NEEDED -I. -I. -I../../.././gcc
    -I../../../../gcc-4.3.1/libgcc -I../../../../gcc-4.3.1/libgcc/.
    -I../../../../gcc-4.3.1/libgcc/../gcc
    -I../../../../gcc-4.3.1/libgcc/../include -DHAVE_CC_TLS -o emutls_s.o
    -MT emutls_s.o -MD -MP -MF emutls_s.dep -DSHARED -fexceptions -c
    .../../../../gcc-4.3.1/libgcc/../gcc/emutls.c
    Assembler: emutls.c
    "/var/tmp//ccboai.Z.s", line 1911 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB2", ".DOT-22"
    "/var/tmp//ccboai.Z.s", line 1920 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB3", ".DOT-23"
    "/var/tmp//ccboai.Z.s", line 1939 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB5", ".DOT-24"
    "/var/tmp//ccboai.Z.s", line 1948 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB6", ".DOT-25"
    "/var/tmp//ccboai.Z.s", line 1957 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB7", ".DOT-26"
    "/var/tmp//ccboai.Z.s", line 1966 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB8", ".DOT-27"
    "/var/tmp//ccboai.Z.s", line 1975 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB16", ".DOT-28"
    "/var/tmp//ccboai.Z.s", line 1984 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB17", ".DOT-29"
    "/var/tmp//ccboai.Z.s", line 1993 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB18", ".DOT-30"
    "/var/tmp//ccboai.Z.s", line 2002 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB23", ".DOT-31"
    "/var/tmp//ccboai.Z.s", line 2011 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB20", ".DOT-32"
    "/var/tmp//ccboai.Z.s", line 2030 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB21", ".DOT-33"
    "/var/tmp//ccboai.Z.s", line 2057 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB19", ".DOT-34"
    "/var/tmp//ccboai.Z.s", line 2086 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB12", ".DOT-35"
    "/var/tmp//ccboai.Z.s", line 2111 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB11", ".DOT-36"
    "/var/tmp//ccboai.Z.s", line 2134 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB10", ".DOT-37"
    "/var/tmp//ccboai.Z.s", line 2157 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB9", ".DOT-38"
    "/var/tmp//ccboai.Z.s", line 2180 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB4", ".DOT-39"
    "/var/tmp//ccboai.Z.s", line 2205 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB22", ".DOT-40"
    "/var/tmp//ccboai.Z.s", line 2238 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB14", ".DOT-41"
    "/var/tmp//ccboai.Z.s", line 2261 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB15", ".DOT-42"
    "/var/tmp//ccboai.Z.s", line 2284 : Warning: Illegal
    subtraction - symbols from different sections: ".LFB13", ".DOT-43"
    # Recursively invoke make in the GCC directory to build any
    # startfiles (for now). We must do this just once, passing
    # it all the GCC_EXTRA_PARTS as simultaneous goal targets,
    # so that rules which cannot execute simultaneously are properly
    # serialized. We indirect through T_TARGET in case any multilib
    # directories contain an equals sign, to prevent make from
    # interpreting any of the goals as variable assignments.
    # We must use cd && make rather than make -C, or else the stage
    # number will be embedded in debug information.
    T=`${PWDCMD-pwd}`/ \
    && cd ../../.././gcc \
    && gmake GCC_FOR_TARGET="/export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include" \
    MULTILIB_CFLAGS="-g -fkeep-inline-functions -m64" \
    T=$T \
    T_TARGET="${T}crtbegin.o ${T}crtend.o ${T}gmon.o" \
    T_TARGET
    gmake[6]: Entering directory `/export/home/drkirkby/build/gcc'
    /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -O2 -O2 -g -g -O2 -DIN_GCC
    -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes
    -Wold-style-definition -isystem ./include -I.
    -I/export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc
    -I../../gcc-4.3.1/gcc
    -I../../gcc-4.3.1/gcc//export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc
    -I../../gcc-4.3.1/gcc/../include -I./../intl
    -I../../gcc-4.3.1/gcc/../libcpp/include -I/usr/local/include
    -I/usr/local/include -I/usr/local/include
    -I../../gcc-4.3.1/gcc/../libdecnumber
    -I../../gcc-4.3.1/gcc/../libdecnumber/dpd -I../libdecnumber -g
    -fkeep-inline-functions -m64 -g0 -finhibit-size-directive
    -fno-inline-functions -fno-exceptions -fno-zero-initialized-in-bss
    -fno-toplevel-reorder -fno-tree-vectorize -fPIC -O2
    -fno-omit-frame-pointer -fno-asynchronous-unwind-tables \
    -c ../../gcc-4.3.1/gcc/crtstuff.c -DCRT_BEGIN \
    -o
    /export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc/crtbegin.o
    /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -O2 -O2 -g -g -O2 -DIN_GCC
    -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes
    -Wold-style-definition -isystem ./include -I.
    -I/export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc
    -I../../gcc-4.3.1/gcc
    -I../../gcc-4.3.1/gcc//export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc
    -I../../gcc-4.3.1/gcc/../include -I./../intl
    -I../../gcc-4.3.1/gcc/../libcpp/include -I/usr/local/include
    -I/usr/local/include -I/usr/local/include
    -I../../gcc-4.3.1/gcc/../libdecnumber
    -I../../gcc-4.3.1/gcc/../libdecnumber/dpd -I../libdecnumber -g
    -fkeep-inline-functions -m64 -g0 -finhibit-size-directive
    -fno-inline-functions -fno-exceptions -fno-zero-initialized-in-bss
    -fno-toplevel-reorder -fno-tree-vectorize -fPIC -O2
    -fno-omit-frame-pointer -fno-asynchronous-unwind-tables \
    -c ../../gcc-4.3.1/gcc/crtstuff.c -DCRT_END \
    -o
    /export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc/crtend.o
    /export/home/drkirkby/build/./gcc/xgcc
    -B/export/home/drkirkby/build/./gcc/
    -B/usr/local/i386-pc-solaris2.11/bin/
    -B/usr/local/i386-pc-solaris2.11/lib/ -isystem
    /usr/local/i386-pc-solaris2.11/include -isystem
    /usr/local/i386-pc-solaris2.11/sys-include -O2 -g -g -O2 -DIN_GCC
    -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes
    -Wold-style-definition -isystem ./include -I.
    -I/export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc
    -I../../gcc-4.3.1/gcc
    -I../../gcc-4.3.1/gcc//export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc
    -I../../gcc-4.3.1/gcc/../include -I./../intl
    -I../../gcc-4.3.1/gcc/../libcpp/include -I/usr/local/include
    -I/usr/local/include -I/usr/local/include
    -I../../gcc-4.3.1/gcc/../libdecnumber
    -I../../gcc-4.3.1/gcc/../libdecnumber/dpd -I../libdecnumber -g
    -fkeep-inline-functions -m64 \
    -c ../../gcc-4.3.1/gcc/config/i386/gmon-sol2.c -o
    /export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc/gmon.o
    Assembler: gmon-sol2.c
    "/var/tmp//ccvPa4s0.s", line 24 : Illegal mnemonic
    Near line: " callq internal_mcount"
    "/var/tmp//ccvPa4s0.s", line 24 : Syntax error
    Near line: " callq internal_mcount"
    "/var/tmp//ccvPa4s0.s", line 24 : Illegal mnemonic
    Near line: " callq internal_mcount"
    gmake[6]: ***
    [/export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc/gmon.o]
    Error 1
    gmake[6]: Leaving directory `/export/home/drkirkby/build/gcc'
    gmake[5]: *** [gcc-extra-parts] Error 2
    gmake[5]: Leaving directory
    `/export/home/drkirkby/build/i386-pc-solaris2.11/amd64/libgcc'
    gmake[4]: *** [multi-do] Error 1
    gmake[4]: Leaving directory
    `/export/home/drkirkby/build/i386-pc-solaris2.11/libgcc'
    gmake[3]: *** [all-multi] Error 2
    gmake[3]: Leaving directory
    `/export/home/drkirkby/build/i386-pc-solaris2.11/libgcc'
    gmake[2]: *** [all-stage1-target-libgcc] Error 2
    gmake[2]: Leaving directory `/export/home/drkirkby/build'
    gmake[1]: *** [stage1-bubble] Error 2
    gmake[1]: Leaving directory `/export/home/drkirkby/build'
    gmake: *** [bootstrap] Error 2
    drkirkby@kingfisher:[~/build] $





  20. Re: gcc build fails "cannot compute suffix of object files"

    On Mon, 09 Jun 2008 23:50:03 +0100, Dave wrote:

    > AGT wrote:
    >> On Mon, 09 Jun 2008 14:25:42 +0100, Dave wrote:
    >>
    >>> Dave wrote:
    >>>> I'm trying to compile gcc 4.3.1 on a Laptop running Solaris Express
    >>>> Community Edition Build 89.
    >>> Looking at my log, I overlooked to mention this line - see full lot
    >>> posted previously.
    >>> xgcc: Internal error: Killed (program cc1)
    >>> Any ideas how to solve it?

    >> You have to read the config.log. As suggested. I just got this "error" too
    >> when bootstrapping in libgcc's configure stage. According to the log
    >> libgmp and libmpfr cannot be found. Which is of course ridiculous as - up
    >> til now anyway - there has NOT been a problem finding the stuff. I had to
    >> start from scratch with MORE configure options, Im adding LDFLAGS.
    >> Eventually this piece of crap will compile ??


    Eventually it did so that worked.
    I just built mplayer from a source snapshot in 32 bit.
    64 Im not so sure yet

    > A look on the 'Successful builds' page for this release of gcc:
    > http://gcc.gnu.org/gcc-4.3/buildstat.html
    > shows only SPARC and then only Solaris 2.6 and 2.8. Nothing about
    > Solaris on x86, or any modern version of Solaris. Clearly the GNU
    > developers do not take portability too seriously.


    And your just starting to figure this out? : >

    > It's hard to believe they would release a version of gcc without testing
    > it properly on at the very least Solaris 10 on x86 and SPARC.


    That would require skill sets they dont seem to possess

    > I suspect the problem in the usual one - people don't make the effort to
    > check that code builds properly on other platforms.


    There is only Penguin OS...
    There is only Intel...
    Gee sounds almost like Mr. Bill all over again huh?

    > I think Sun could do a lot here though, by making Sun hardware available
    > to software developers for free by remote access. If Sun did that, there
    > might be a few more gcc developers who would take the trouble to test
    > the code on Solaris.


    Seem to recall there is an initiative to test your software/equipment on
    Sun equipment. i dont write software so i never followed up.

    Fact is there is no excuse as you can get a SPARC box for next to nothing
    on eBay for a test platform and theres VMWare or Virtual Box for FeeCees.

    > HP offer the TestDrive site
    > http://www.testdrive.hp.com/
    > where anyone can get an account on a range of HP machines to test
    > software. I've used it myself several times.
    > When I developed 'ATLC'
    > http://atlc.sourceforge.net/
    > I made sure it compiled on pretty much anything. Solaris, Linux, AIX,
    > HP-UX, Tru64, IRIX, UNICOS, NetBSD, OpenBSD, FreeBSD etc etc. The
    > biggest hassle was an old Cray, where sizeof(short) == 8.


    Its really not that hard to write portable C
    but it might be harder on Linux with their header files
    like what is it and the like.
    I bet I wont find that header on my 5120 when it arrives..
    Unless I put Linux on a Logical Domain..

    > I'm trying to help with a port of the maths program Sage
    > http://www.sagemath.org/
    > to Solaris and have made a Sun Ultra 60 available to a couple of
    > developers, so they can check their changes on Solaris. Hopefully there
    > will be a Solaris port of Sage within days, or at the most weeks.


    Well thats good of you. kudos


+ Reply to Thread
Page 1 of 3 1 2 3 LastLast