OOM errors, was Re: panic: Assertion td->td_lock == TDQ_LOCKPTR(tdq) failed at /usr/src/sys/kern/sched_ule.c:2137

[bob prohaska]

> 
> > On Sat, Jun 02, 2018 at 09:14:18AM -0600, Warner Losh wrote:
> > >
> > > Do you have r334508? It fixes the OOM falsely triggering.
> > >
> 
The kernel is now up to 334538, but I'm still seeing OOM kills, like this:

Jun  4 00:01:57 www kernel: pid 27824 (llvm-tblgen), uid 0, was killed: out of swap space
Jun  4 00:01:57 www kernel: pid 27828 (llvm-tblgen), uid 0, was killed: out of swap space
Jun  4 00:01:57 www kernel: pid 27832 (llvm-tblgen), uid 0, was killed: out of swap space
Jun  4 00:01:57 www kernel: pid 27830 (llvm-tblgen), uid 0, was killed: out of swap space
Jun  4 00:01:57 www kernel: pid 27610 (make), uid 0, was killed: out of swap space


After restarting the -j4 buildworld with -DNO_CLEAN it appeared that
swap use peaked around 129 MB. Four instances of llvm-tblgen entered
state pfault at roughly the same time  swap use peaked and then two
were killed. Apparently the remaining pair died in sympathy.

Have you any idea what I might be doing wrong?

Thanks for reading,

bob prohaska
`