[Bug 281560] gve (4) uma deadlock during high tcp throughput

From: <bugzilla-noreply_at_freebsd.org>
Date: Fri, 27 Sep 2024 14:59:19 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=281560

--- Comment #17 from Konstantin Belousov <kib@FreeBSD.org> ---
(In reply to Mark Johnston from comment #16)
I doubt that system would stay silent about a CPU with disabled interrupts,
our IPI code does not tolerate such condition.

In fact, I asked about pcpu because for me it more sounds as if idle thread
does not exit an idle loop.  Note that this stuff occurring in the VM, and
we e.g. MWAIT emulation might be tricky.  Would the CPU in question sit in
idle thread, I suggest to change the idling method either to hlt or loop.

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