https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107590
--- Comment #3 from Sergey Fedorov <vital.had at gmail dot com> --- (In reply to Andrew Pinski from comment #2) > >Reason: 259 at address: 0x00003109 > > Yes that does seem like an alignment disagreement. > > I suspect the code is broken for allocation and it is allocating unaligned > structs. The code in Boost shared pointer tests looks pretty simple. Since the bus error happens on those, no need to look into i2pd case. I don’t really know how alignment works there, but hopefully someone can look at that. But the error was reproducible on several tests plus i2pd itself. All fixed by switching to Boost own atomics. I am not qualified to judge if the code is broken, but this looks like it is worth investigation. > Also inside gdb can you do the following: > > disassemble $pc-0x10 $pc+0x10 > info registers I could try that tomorrow, provided an ancient GBD we have on PPC supports that.
