@trasz could you please say something about this?

Found a race between sys_exit() -> exit1() -> thread_exit() and sys_wait() -> proc_reap().
Slightly moving down PROC_SUNLOCK() in thread_exit() to protect RACCT_RT calculations from proc_reap() destroying p->p_racct.

For example it is possible to share file descriptor tables, and one of the processes may not be encumbered by the jail.

Found same function in lib/libkvm/kvm_proc.c, patch it too.

As for 64/64 divides on 32bit platform: in fact, (nearly) all 32-bit CPUs have a 32-bit tickrate (<4.2GHz), and modern compilers/libraries seems to use 64/32 math in runtime for such a case in 32-bit mode (__udivdi3 function).

I didn't done the test by myself, but this is trivial integer arithmetic.

Is it possible to find out how this was tested?

Is "firk <firk@cantconnect.ru>" the desired author for the git commit?

simplify things according to replies