Bug#851412: FTBFS on mips*, tests fail with SIGSEGV

[Michael Biebl]

Source: systemd
Version: 232-10
Severity: serious
User: debian-mips at lists.debian.org
Usertags: mips-port

Hi mips porters,

the latest upload of systemd FTBFS on mips* [1]. Contrary to what the
buildd status page shows, mips64el is also affected. The build chroot on
mips64el did not have a /etc/machine-id, which is why the test suite was
non-fatal [2]. If you install systemd in the chroot (which sets up
/etc/machine-id) mips64el fails exactly like mips and mipsel, i.e. the
following two tests segfault:

../build-aux/test-driver: line 107: 26379 Segmentation fault   "$@" > $log_file 2>&1
FAIL: test-bus-marshal

../build-aux/test-driver: line 107: 27445 Segmentation fault   "$@" > $log_file 2>&1
FAIL: test-libsystemd-sym

Building the package on a porter pox (eller: mipsel, mips64el, minkus:
mips), the problem is 100% reproducible.

dmesg shows
[7368858.796075] 
                 do_page_fault(): sending SIGSEGV to test-bus-marsha for invalid read access from 000000000064658b
[7368858.806215] epc = 0000000077e19058 in
[7368858.829057]  libsystemd-shared-232.so[77dda000+180000]
[7368858.832726] ra  = 0000000077e19078 in
[7368858.845049]  libsystemd-shared-232.so[77dda000+180000]

[7368895.490458] 
                 do_page_fault(): sending SIGSEGV to test-libsystemd for invalid read access from 000000000064658b
[7368895.500588] epc = 0000000077b85f44 in
[7368895.524124]  libsystemd.so.0.17.0[77b77000+82000]
[7368895.527490] ra  = 0000000077b85f64 in
[7368895.536107]  libsystemd.so.0.17.0[77b77000+82000]



The initial test builds on the porter machines were using
binutils_2.27.90.20170113-1. As I suspected a binutils problem, I
upgraded to the recently uploaded 
binutils_2.27.90.20170114-1. Building in stretch chroot though was
successful. So my observation is:

binutils_2.27.90.20170114-1: fails
binutils_2.27.90.20170113-1: fails
binutils_2.27.51.20161220-1: ok

So it looks like a mips related regression in binutils at a first
glance. In the past, systemd built reasonably reliably on mips* [3].


Dear mips porters, it would be great if you can have a look. Can you
confirm these findings or is there something else causing these issues?

I've also CCed the binutils maintainers so they are aware of this
problem. Maybe they have an idea what's going wrong.

Regards,
Michael


[1] https://buildd.debian.org/status/package.php?p=systemd&suite=sid
[2] https://anonscm.debian.org/cgit/pkg-systemd/systemd.git/tree/debian/rules#n357
[3] https://buildd.debian.org/status/logs.php?pkg=systemd&arch=mipsel
-- Package-specific info:

-- System Information:
Debian Release: stretch/sid
  APT prefers unstable-debug
  APT policy: (500, 'unstable-debug'), (500, 'unstable'), (200, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.8.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages systemd is related to:
pn  dracut           <none>
ii  initramfs-tools  0.126
ii  udev             232-10

-- no debconf information