Bruce Dubbs wrote:
Bruce Dubbs wrote:
Douglas R. Reno wrote:
On Thu, Aug 4, 2016 at 6:03 PM, Bruce Dubbs <[email protected]> wrote:

I started to test both binutils-2.27 and glibc-2.24 today and got a crash
at binutils-pass2.  I then backed off and tried binutils-2.27 with the
existing glibc and still got the crash.  Trying the new glibc and the old
binutils was OK.

The binutils-pass2 error is in the configure section:

configure:4496: checking whether the C compiler works
configure:4505: ./a.out
../configure: line 4507: 31596 Segmentation fault      ./$ac_file
If you meant to cross compile, use `--host'.

I'm still investigating, but if anyone has a suggestion, let me know.
My next step will be to see what valid options can be used for --host.

Checking their 2.27 branch in git, all I have seen is:

https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commit;h=dccb2ff522d2afba77301a046b37949356c29f88



However, I am not sure if that is the case without more context. I can
search through all commits since it was released and see if there is a fix
present there.

Just now I was able to get binutils-pass2 to build with adding
--host=$LFS_TGT like we do in  Libstdc++-6.1.0

That was just a preliminary indication.  I'll try a full build with that.
2.26.1 an try a
Did not work.  I did get a lot further but got an error:

  libtool: compile:  x86_64-lfs-linux-gnu-gcc -DHAVE_CONFIG_H -I.
-I../../ld -I. -I../../ld -I../bfd -I../../ld/../bfd -I../../ld/../include
-g -O2 -DENABLE_PLUGINS -DLOCALEDIR=\"/tools/share/locale\" -g -O2 -g -O2
-MT libldtestplug_la-testplug.lo -MD -MP -MF
.deps/libldtestplug_la-testplug.Tpo -c ../../ld/testplug.c  -fPIC -DPIC -o
.libs/libldtestplug_la-testplug.o
x86_64-lfs-linux-gnu-gcc: internal compiler error: Segmentation fault
(program as)

So the assembler is dropping a segmentation error.  I'm going to drop back
to binutils-2.26.1 and try a full build.

Wow. These latest changes are really causing a problem. Actually the only file I changed was glibc, but is caused 214 unexpected errors in gcc and required a sed in sysklogd.

It's getting late and we'll need to investigate.  I'll look at it tomorrow.

All the gcc failures look like:

FAIL: c-c++-common/asan/asan-interface-1.c   -O0  execution test

with different gcc switches.  The most common is in the files like
c-c++-common/asan/bitfield-?.c  where the ? is 1, 2, 3, 4, 5
c-c++-common/asan/clone-test-1.c
c-c++-common/asan/global-overflow-1.c
c-c++-common/asan/halt_on_error-1.c
c-c++-common/asan/heap-overflow-1.c
c-c++-common/asan/memcmp-1.c
c-c++-common/asan/misalign-1.c
c-c++-common/asan/null-deref-1.c

  -- Bruce

--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Reply via email to