On 21.10.19 13:36, Roger Pau Monné wrote:
On Mon, Oct 21, 2019 at 12:52:10PM +0200, Jürgen Groß wrote:
On 21.10.19 11:51, Sergey Dyasli wrote:
Hello,
While testing pv-shim from a snapshot of staging 4.13 branch (with core-
scheduling patches applied), some sort of scheduling issues were uncovered
which usually leads to a guest lockup (sometimes with soft lockup messages
from Linux kernel).
This happens more frequently on SandyBridge CPUs. After enabling
CONFIG_DEBUG in pv-shim, the following assertions failed:
Null scheduler:
Assertion 'lock == get_sched_res(i->res->master_cpu)->schedule_lock'
failed at ...are/xen-dir/xen-root/xen/include/xen/sched-if.h:278
(full crash log: https://paste.debian.net/1108861/ )
Credit1 scheduler:
Assertion 'cpumask_cycle(cpu, unit->cpu_hard_affinity) == cpu' failed at
sched_credit.c:383
(full crash log: https://paste.debian.net/1108862/ )
I'm currently investigation those, but would appreciate any help or
suggestions.
Hmm, I think that pv_shim_cpu_up() shouldn't do the vcpu_wake(), but the
caller.
Does the attached patch help?
Juergen
From 068ea0419d1a67e967b9431ed11e24b731cd357f Mon Sep 17 00:00:00 2001
From: Juergen Gross <jgr...@suse.com>
Date: Mon, 21 Oct 2019 12:28:33 +0200
Subject: [PATCH] xen/shim: fix pv-shim cpu up/down
Calling vcpu_wake() from pv_shim_cpu_up() is wrong as it is not yet
sure that the correct scheduler has taken over the cpu.
I'm not sure why this is wrong, the scheduler should be capable of
handling 2 vCPUs on a single pCPU while the new pCPU is brought
online?
Oh, right, I made some false assumptions.
This patch is pure nonsense.
Juergen
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
