gcc compile failure (ld) for armv5tel F15

[Daniel Drake]

Hi fossjon and others,

Here is a gdb backtrace from the ld crash from when building
gcc-4.6.0-6.ct001.fc15.src.rpm


/usr/bin/ld --build-id --no-add-needed --eh-frame-hdr --hash-style=gnu
-export-dynamic -dynamic-linker /lib/ld-linux.so.3 -X -m
armelf_linux_eabi -o cc1
/usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1/../../../crt1.o
/usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1/../../../crti.o
/usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1/crtbegin.o
-L/usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1
-L/usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1/../../.. c-lang.o
c-family/stub-objc.o attribs.o c-errors.o c-decl.o c-typeck.o
c-convert.o c-aux-info.o c-objc-common.o c-parser.o tree-mudflap.o
c-family/c-common.o c-family/c-cppbuiltin.o c-family/c-dump.o
c-family/c-format.o c-family/c-gimplify.o c-family/c-lex.o
c-family/c-omp.o c-family/c-opts.o c-family/c-pch.o
c-family/c-ppoutput.o c-family/c-pragma.o c-family/c-pretty-print.o
c-family/c-semantics.o c-family/c-ada-spec.o arm-c.o cc1-checksum.o
main.o libbackend.a ../libcpp/libcpp.a ../libdecnumber/libdecnumber.a
../libcpp/libcpp.a ../libiberty/libiberty.a
../libdecnumber/libdecnumber.a -lmpc -lmpfr -lgmp -ldl -lz -lgcc
--as-needed -lgcc_s --no-as-needed -lc -lgcc --as-needed -lgcc_s
--no-as-needed /usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1/crtend.o
/usr/lib/gcc/armv5tel-redhat-linux-gnueabi/4.5.1/../../../crtn.o



Program received signal SIGSEGV, Segmentation fault.
sha1_process_block (buffer=<value optimized out>, len=<value optimized out>,
    ctx=0x6a578500) at ./sha1.c:319
319	  while (words < endp)
Missing separate debuginfos, use: debuginfo-install
glibc-2.13-2.1.armv5tel libgcc-4.5.1-5.fc14.1.armv5tel
zlib-1.2.5-2.fc14.armv5tel
(gdb)
(gdb) bt
#0  sha1_process_block (buffer=<value optimized out>,
    len=<value optimized out>, ctx=0x6a578500) at ./sha1.c:319
#1  0x000348c8 in sha1_process_bytes (buffer=0x43ef2068, len=42803016,
    ctx=0xbefff430) at ./sha1.c:245
#2  0x4006fc44 in bfd_elf32_checksum_contents (abfd=0x92530,
    process=0x34808 <sha1_process_bytes>, arg=0xbefff430) at elfcode.h:1182
#3  0x0002d090 in gldarmelf_linux_eabi_write_build_id_section (abfd=0x90558)
    at earmelf_linux_eabi.c:1394
#4  0x400790d8 in _bfd_elf_write_object_contents (abfd=0x90558) at elf.c:5321
#5  0x4004301c in bfd_close (abfd=0x90558) at opncls.c:699
#6  0x000229fc in main (argc=10000, argv=0x43) at ./ldmain.c:497

Daniel

[Daniel Drake]

On Fri, Sep 23, 2011 at 12:09 AM, Daniel Drake <dsd(a)laptop.org&gt; wrote:
>  Hi fossjon and others,
>
>  Here is a gdb backtrace from the ld crash from when building
>  gcc-4.6.0-6.ct001.fc15.src.rpm 
That crash was with binutils-2.21.51.0.6-2.fc15
I also tested binutils-2.21.51.0.6-7.fc15 - same crash

Then I built binutils-2.21.53.0.2-1.fc17 for F15 armv5tel, and that
one doesn't fail. Well, gcc is still building (takes 2 days if I
remember right), but previously it failed on the first 'cc1' link
which is now long gone.

I'm going to try some intermediate versions in order to narrow down
the set of possible fixes, which we could then backport to F15.

Daniel

[Daniel Drake]

On Fri, Sep 23, 2011 at 7:33 PM, Daniel Drake <dsd(a)laptop.org&gt; wrote:
>  Then I built binutils-2.21.53.0.2-1.fc17 for F15 armv5tel, and that
>  one doesn't fail. Well, gcc is still building (takes 2 days if I
>  remember right), but previously it failed on the first 'cc1' link
>  which is now long gone. 
It failed at the very very end of the build. We hit
https://bugzilla.redhat.com/show_bug.cgi?id=697685.

I am now trying again with gcc-4.6.1-9.fc15 modified to exclude cloog
and java (like ctyler's build).

And I added a note here:
https://fedoraproject.org/wiki/Architectures/ARM/Fedora15_HardFP_Bootstra...

This is important - the koji F15 build should not bother with any gcc
version older than gcc-4.6.1-9.fc15. Don't waste 3 days of your life
to only hit the same error at the very end, or to end up with a gcc
that produces broken libraries.



Going back to the ld crash issue...

>  I'm going to try some intermediate versions in order to narrow
> down
>  the set of possible fixes, which we could then backport to F15. 
I found the point of breakage.
binutils-2.21.51.0.8-3.fc16 reproduces the ld crash.
binutils-2.21.51.0.9-1.fc16 works fine.

Here is the diff between 2.21.51.0.8 and 2.21.51.0.9 :
http://dev.laptop.org/~dsd/20110924/binutils-2.21.51.0.8-to-binutils-2.21...
40k lines bedtime reading. The fix must be in there somewhere -
anything jump out at anyone?

Thanks,
Daniel

[Daniel Drake]

OK, I think I have a winning combination:

gcc-4.6.1-9.fc15.dsd1 from
http://dev.laptop.org/~dsd/20110924/gcc-4.6.1-9.fc15.dsd1.src.rpm
This is gcc-4.6.1-9.fc15 modified (for bootstrapping purposes) to
remove the circular dependency on cloog and to disable the java build
(like ctyler did). Using the new version 4.6.1-9 is important to avoid
#697685 and #733549.


I figured out the binutils issue and explained it here:
https://bugzilla.redhat.com/show_bug.cgi?id=741053

Here is a version of binutils-2.21.51.0.6-7.fc15 which has the
suggested patch included (already built):
http://dev.laptop.org/~dsd/20110924/binutils-2.21.51.0.6-7.fc15.1/
I will followup to make sure this gets included in Fedora soon.


Put those 2 together and you'll get a working gcc out at the end. My
builder is busy so I can't kick off the gcc build yet (but I'm
confident it will work). fossjon, can you get this underway on one of
your systems?

Thanks,
Daniel

p.s. Whats the next blocking F15 armv5tel issue I could contribute to?