[Bug bootstrap/61078] New: ESA mode bootstrap failure since r209897

[krebbel at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61078

            Bug ID: 61078
           Summary: ESA mode bootstrap failure since r209897
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: bootstrap
          Assignee: unassigned at gcc dot gnu.org
          Reporter: krebbel at gcc dot gnu.org

Regression hunt indicates that r209897 might have caused a bootstrap failure on
s390 ESA mode:

Bootstrap dies reproducable with a segfault in bash:
...
make install-leaf DESTDIR=../.././gcc \
  slibdir= libsubdir= MULTIOSDIR=.
make[4]: Entering directory
`/build/patched/gcc-4.10.0-build/s390-ibm-linux-gnu/libgcc'
/bin/sh /build/patched/gcc/libgcc/../mkinstalldirs ../.././gcc
/usr/bin/install -c -m 644 libgcc_eh.a ../.././gcc/
chmod 644 ../.././gcc/libgcc_eh.a
ranlib ../.././gcc/libgcc_eh.a
/bin/sh /build/patched/gcc/libgcc/../mkinstalldirs ../.././gcc;
/usr/bin/install -c -m 644 ./libgcc_s.so.1 ../.././gcc/libgcc_s.so.1; rm -f
../.././gcc/libgcc_s.so; ln -s libgcc_s.so.1 ../.././gcc/libgcc_s.so
/bin/sh /build/patched/gcc/libgcc/../mkinstalldirs ../.././gcc
make[4]: *** [install-leaf] Segmentation fault
make[4]: Leaving directory
`/build/patched/gcc-4.10.0-build/s390-ibm-linux-gnu/libgcc'
make[3]: *** [all] Error 2
make[3]: Leaving directory
`/build/patched/gcc-4.10.0-build/s390-ibm-linux-gnu/libgcc'
make[2]: *** [all-stage2-target-libgcc] Error 2
make[2]: Leaving directory `/build/patched/gcc-4.10.0-build'
make[1]: *** [stage2-bubble] Error 2
make[1]: Leaving directory `/build/patched/gcc-4.10.0-build'
make: *** [all] Error 2


[pid  1082] vfork(Process 1088 attached
 <unfinished ...>
[pid  1088] rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
[pid  1088] prlimit64(0, RLIMIT_STACK, {rlim_cur=8192*1024,
rlim_max=RLIM64_INFINITY}, NULL) = 0
[pid  1088] execve("/bin/sh", ["/bin/sh", "-c", "subdirs='testsuite';
\\\ntarget=`e"...], [/* 172 vars */] <unfinished ...>
[pid  1082] <... vfork resumed> )       = 1088
[pid  1082] rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
[pid  1082] wait4(-1,  <unfinished ...>
[pid  1088] <... execve resumed> )      = 0
....
[pid  1088] --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR,
si_errno=1825521056, si_addr=0} ---
[pid  1088] +++ killed by SIGSEGV +++
[pid  1082] <... wait4 resumed> [{WIFSIGNALED(s) && WTERMSIG(s) == SIGSEGV}],
0, NULL) = 1088
[pid  1082] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=1088,
si_status=SIGSEGV, si_utime
=0, si_stime=0} ---
[pid  1082] sigreturn() (mask [])       = 1088

I've verified that reverting the patch "fixes" the problem. I haven't figured
out yet how the patch is related to the problem. Perhaps the execve arguments
for /bin/sh somehow get messed up with the patch?!