[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 8 Aug 2023 12:20:29 +0100
From: Andrew Cooper <andrew.cooper3@...rix.com>
To: Juergen Gross <jgross@...e.com>,
Thomas Gleixner <tglx@...utronix.de>,
LKML <linux-kernel@...r.kernel.org>
Cc: x86@...nel.org, Tom Lendacky <thomas.lendacky@....com>,
Arjan van de Ven <arjan@...ux.intel.com>,
Huang Rui <ray.huang@....com>,
Dimitri Sivanich <dimitri.sivanich@....com>,
Michael Kelley <mikelley@...rosoft.com>,
Sohil Mehta <sohil.mehta@...el.com>,
K Prateek Nayak <kprateek.nayak@....com>,
Kan Liang <kan.liang@...ux.intel.com>,
Zhang Rui <rui.zhang@...el.com>,
"Paul E. McKenney" <paulmck@...nel.org>,
Feng Tang <feng.tang@...el.com>,
Andy Shevchenko <andy@...radead.org>
Subject: Re: [patch 00/53] x86/topology: The final installment
On 08/08/2023 8:40 am, Juergen Gross wrote:
> Tested on an Intel system with Xen:
>
> - PV dom0 is working fine. I couldn't test physical cpu hotplug, but
> removing
> and then re-adding vcpus to dom0 worked.
It turns out that physical CPU hotplug with XenPV is broken in at least
two ways.
It's dom0 (not Xen) that gets the hot-unplug event, after which the Xen
code in Linux succumbs to a preempt-check failure while trying to
offline the vCPU that aliases the pCPU wanting to go offline.
~Andrew
Powered by blists - more mailing lists