Michael,
Thanks for the comment. Your instructions were very clear and the
process was easy, if a little tedious.
I hunted up the qemu-devel mailing list, found your ping and looked at
the reply. There is a link there to a patch you proposed at
https://gitlab.com/qemu-project/qemu/-/issues/1628#note_2144606625. I
reverted to v9.1.1, applied your patch, ran the VM for over an hour,
shut it down and ran it again, and lo and behold, no more freezing. I
won't pretend I understand what it does nor why my machine freezes and
yours don't, but it fixes the issue on my machine. If you figure out
what's going on, I'd love to know.
Hank
On 10/30/24 17:52, Michael Tokarev wrote:
Hank, you did an excellent job here, much more than I'd expect
given the previous conversation.
One of the very first thoughts about this is that the test
for the bisection is a flaky one, since you have to wait for
"some" time for the freeze to happen, and it isn't deterministic.
So the first suspect was the faulty bisection. But you has proven
your point now, much better than I'd thought.
I pinged the patch in question on the mailinglist. It received
at least single reply - also suggesting a possible flaky bisection
as a first possibility. Nope, this is definitely not it.
I'll do some more pinging tomorrow.
Thank you for a really excellent job!
/mjt
--
Hank Knox, FRSC
Schulich School of Music of
McGill University (retired)
Montreal, QC
