Hi Dave, The system works significantly more stable with VMs running and patched vmm.c from -current. In four days I have encountered only one hang which lead to shutdown using power button. Hope it was independent issue.
Martin ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ On Friday, May 21, 2021 6:59 AM, Martin <[email protected]> wrote: > Hi Dave, > > Following your advice I've rebuilt the kernel from -current with latest vmm.c > patches. > > For last 24hrs since reboot with -current kernel - no issues. I will let you > know if any. > > Martin > > ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ > On Thursday, May 20, 2021 11:06 AM, Dave Voutila [email protected] wrote: > > > Martin writes: > > > > > The hang mostly happened when VM guest run browser or any network > > > activity (repository updates etc). > > > The bad thing I can't debug it because the host system hangs completely. > > > > We'd need ddb backtrace output with register state and details of where > > the panic or fault is to make any diagnosis. You'd need to be outside X > > and at the main console to see it during a panic. Without that info I > > can only speculate this is related to something recently fixed in > > -current for AMD hosts. > > I suggest either trying a -current snapshot or building a custom > > 6.9-stable kernel with the patch supplied recently on bugs@: > > https://marc.info/?l=openbsd-bugs&m=162075185720480&w=2 > > -dv
