[Bug 219399] System panics after several hours of 14-threads-compilation orgies using poudriere on AMD Ryzen...

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=219399

--- Comment #59 from Mark Millard <markmi at dsl-only.net> ---
(In reply to Don Lewis from comment #58)

Good to know that 11.x variants (likely non-debug) and head (12) debug
kernels both get the problems.

Unfortunately the MSI X370 Gaming Pro Carbon 1800X context that I had access
to is unavailable: a BIOS update (from 7A32v15 to 7A32v17) turned it into a
brick. I've no clue if eventual access to another will be as well behaved.

I had no problems but ran FreeBSD (head but non-debug/non-invarints) in a
VirtualBox virtual machine under Windows 10 Pro.

It would be interesting for someone with a known workload that
fails on known hardware could try the workload under a virtual machine
under, say, Windows 10, on the example hardware in order to see if the
problem is repeatable where the OS really in control is a primary one for
the X370 board makers.

If the problem does not occur then the implication would be some
difference in the handling of things makes the difference in the
overall result. (Not that I have a clue what it would then be that
made the difference.)

If the problem instead repeated, that too would be interesting and
might be of more interest to the board makers: crashes under a
primary OS.

-- 
You are receiving this mail because:
You are the assignee for the bug.