Re: rust from 2018Q3 fails on i386

[Havard Eidnes <he%NetBSD.org@localhost>]

> "Dead lock detected" is the cause of failure.

It's a heisenbug.  The error message comes from our ld.elf_so.

I've yet to have explained to me what could trigger this.  It
could be (I'm guessing here) time-parallel resolution of one or
multiple dynamic symbols by more than one thread after multiple
threads have been created...

Regards,

- Håvard