This is the mail archive of the libc-alpha@sources.redhat.com mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Fixing MIPS target under GLIBC....


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



On Wed, 14 Mar 2001, Steven J. Hill wrote:
> FAILED BUILD
> ------------
> binutils-latest-cvs
> gcc-latest-cvs
> glibc-latest-cvs AND glibc-2.2.2
As far as I understand this, latest CVS and glibc-2.2.2 have two different
problems.

> I can successfully build binutils and gcc from CVS for MIPS with
> no problems. I then try to build the latest glibc from CVS and get
> the following (I posted this previously on the mailing list):

> make[2]: Entering directory `/data/cvs/libc/iconv'
> mipsel-linux-gcc -nostdlib -nostartfiles -o
> /data/build/build-glibc/iconv/iconv_prog
> -Wl,-dynamic-linker=/opt/toolchains/mips/mipsle/mipsel-linux//lib/ld.so.1
> /data/build/build-glibc/csu/crt1.o /data/build/build-glibc/csu/crti.o
> `mipsel-linux-gcc --print-file-name=crtbegin.o`
> /data/build/build-glibc/iconv/iconv_prog.o
> /data/build/build-glibc/iconv/iconv_charmap.o
> /data/build/build-glibc/iconv/charmap.o
> /data/build/build-glibc/iconv/charmap-dir.o
> /data/build/build-glibc/iconv/linereader.o
> /data/build/build-glibc/iconv/dummy-repertoire.o
> /data/build/build-glibc/iconv/simple-hash.o
> /data/build/build-glibc/iconv/xstrdup.o /data/build/build-glibc/iconv/xmalloc.o
> -Wl,-rpath-link=/data/build/build-glibc:/data/build/build-glibc/math:/data/build/build-glibc/elf:/data/build/build-glibc/dlfcn:/data/build/build-glibc/nss:/data/build/build-glibc/nis:/data/build/build-glibc/rt:/data/build/build-glibc/resolv:/data/build/build-glibc/crypt:/data/build/build-glibc/linuxthreads
> /data/build/build-glibc/libc.so.6 /data/build/build-glibc/libc_nonshared.a -lgcc
> `mipsel-linux-gcc --print-file-name=crtend.o` /data/build/build-glibc/csu/crtn.o
> /opt/toolchains/mips/mipsle/lib/gcc-lib/mipsel-linux/3.1/libgcc.a(__main.o): In
> function `no symbol':
> /data/cvs/gcc/gcc/libgcc2.c(.text+0x18c): undefined reference to `atexit'
> collect2: ld returned 1 exit status
> make[2]: *** [/data/build/build-glibc/iconv/iconv_prog] Error 1
> make[2]: Leaving directory `/data/cvs/libc/iconv'
> make[1]: *** [iconv/others] Error 2
> make[1]: Leaving directory `/data/cvs/libc'
> make: *** [all] Error 2
Same here. Looking through libc-hack mailing list archives I found that
there were ome changes to atexit, and related makefiles or something very
resently. First post related to this is at 03/03/2001,
Subject: libanl+atexit+more

> Next I tried using the same binutils and gcc from CVS to build GLIBC-2.2.2
> release and get this error during the the configure stage:
>
> sjhill@beast:/data/build/build-glibc2$ BUILD_CC=gcc CC=mipsel-linux-gcc
> AR=mipsel-linux-ar AS=mipsel-linux-as RANLIB=mipsel-linux-ranlib
> ../../looking/glibc-2.2.2/configure
> --prefix=/opt/toolchains/mips/mipsle/mipsel-linux/ mipsel-linux
> --build=i686-pc-linux-gnu --enable-add-ons --with-elf --disable-profile
> --with-headers=/opt/toolchains/mips/mipsle/mipsel-linux/include/
> creating cache ./config.cache
> checking host system type... mipsel-unknown-linux-gnu
> checking sysdep dirs... sysdeps/mips/elf linuxthreads/sysdeps/unix/sysv/linux
> linuxthreads/sysdeps/pthread linuxthreads/sysdeps/unix/sysv
> linuxthreads/sysdeps/unix linuxthreads/sysdeps/mips sysdeps/unix/sysv/linux/mips
> sysdeps/unix/sysv/linux sysdeps/gnu sysdeps/unix/common sysdeps/unix/mman
> sysdeps/unix/inet sysdeps/unix/sysv sysdeps/unix/mips sysdeps/unix sysdeps/posix
> sysdeps/mips/mipsel sysdeps/mips/fpu sysdeps/mips sysdeps/wordsize-32
> sysdeps/ieee754/flt-32 sysdeps/ieee754/dbl-64 sysdeps/ieee754
> sysdeps/generic/elf sysdeps/generic
> checking for a BSD compatible install... /usr/bin/install -c
> checking whether ln -s works... yes
> checking for pwd... /bin/pwd
> checking build system type... i686-pc-linux-gnu
> checking for mipsel-linux-gcc... mipsel-linux-gcc
> checking version of mipsel-linux-gcc... 3.1, bad
> checking for gnumake... no
> checking for gmake... gmake
> checking version of gmake... 3.79.1, ok
> checking for gnumsgfmt... no
> checking for gmsgfmt... no
> checking for msgfmt... msgfmt
> checking version of msgfmt... 0.10.35, ok
> checking for makeinfo... makeinfo
> checking version of makeinfo... 4.0, ok
> checking for gsed... no
> checking for sed... sed
> checking version of sed... 3.02, ok
> checking whether the C compiler (mipsel-linux-gcc  ) works... no
> checking whether the C compiler (mipsel-linux-gcc  ) is a cross-compiler...
> (cached) yes
> checking whether we are using GNU C... yes
> checking build system type... i686-pc-linux-gnu
> checking for gcc... gcc
> checking how to run the C preprocessor... /lib/cpp
> checking whether
> /opt/toolchains/mips/mipsle/lib/gcc-lib/mipsel-linux/3.1/../../../../mipsel-linux/bin/as
> is GNU as... yes
> checking whether
> /opt/toolchains/mips/mipsle/lib/gcc-lib/mipsel-linux/3.1/../../../../mipsel-linux/bin/ld
> is GNU ld... yes
> checking for mipsel-linux-mig... mig
> configure: error:
> *** Some critical program is missing or too old.

Same here. It seems that test whether GCC is working or not
inappropriately fails. I think it is GLIBC configure problem, which is
fixed in latest CVS.

All of above is just my guessing, since I currently neither have time, nor
patience to dive into configure and Makefile hell of glibc & gcc


	Ilya.
/*
 * --IV--
 */

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: pgpenvelope 2.9.0 - http://pgpenvelope.sourceforge.net/

iEYEARECAAYFAjqvu2oACgkQtKh84cA8u2nmUQCfewhbv0naWbjIfK9DXyw49hT2
JNUAnjYVNKYyaB1NpyL4pvpCTQADYWQ0
=fwEp
-----END PGP SIGNATURE-----


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]