https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80236
Jonathan Wakely <redi at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|WAITING |UNCONFIRMED Ever confirmed|1 |0 --- Comment #9 from Jonathan Wakely <redi at gcc dot gnu.org> --- (In reply to Dominik Schmidt from comment #8) > Do you still need a regular gdb backtrace without asan? Looks the same to > me, but I can provide it anyway. This is OK. > `ldd /tmp/crashTest ` prints: > linux-vdso.so.1 (0x7efa3000) > libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x76e10000) > libm.so.6 => /lib/libm.so.6 (0x76d8f000) > libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x76d63000) > libc.so.6 => /lib/libc.so.6 (0x76c27000) > /lib/ld-linux-armhf.so.3 (0x76f51000) > > Yes, I'm pretty sure this is the correct libstdc++. That was confirmed on IRC: lrwxrwxrwx 1 root root 19 Feb 28 18:29 /usr/lib/libstdc++.so.6 -> libstdc++.so.6.0.22 I don't see anything obviously wrong in the libstdc++ code, and can't reproduce this on x86_64, so it might be a misoptimization.