NetBSD Problem Report #55978

From www@netbsd.org  Fri Feb  5 14:33:03 2021
Return-Path: <www@netbsd.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 4EA121A9217
	for <gnats-bugs@gnats.NetBSD.org>; Fri,  5 Feb 2021 14:33:03 +0000 (UTC)
Message-Id: <20210205143301.954101A9239@mollari.NetBSD.org>
Date: Fri,  5 Feb 2021 14:33:01 +0000 (UTC)
From: prlw1@cam.ac.uk
Reply-To: prlw1@cam.ac.uk
To: gnats-bugs@NetBSD.org
Subject: panic: HYPERVISOR_mmu_update failed
X-Send-Pr-Version: www-1.0

>Number:         55978
>Category:       port-xen
>Synopsis:       panic: HYPERVISOR_mmu_update failed
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    port-xen-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Feb 05 14:35:00 +0000 2021
>Last-Modified:  Sun Apr 11 13:30:01 +0000 2021
>Originator:     Patrick Welche
>Release:        9.99.79 everything from 5 Feb 2021
>Organization:
>Environment:
>Description:
As xentools now builds (thanks for fixing PR 55888), I tried to boot a -current XEN_DOM0 with the latest sysutis xen{kernel,tools}413, but no joy (inability to detect the root partition is just an aside - non-XEN boot works fine):


(XEN) Xen version 4.13.2nb5 (prlw1@) (gcc (nb1 20200907) 9.3.0) debug=n  Fri Feb  5 09:20:25 GMT 2021
...
(XEN) AMD-Vi: IOMMU 0 Enabled.
(XEN) I/O virtualisation enabled
...
(XEN) HVM: ASIDs enabled.
(XEN) SVM: Supported advanced features:
(XEN)  - Nested Page Tables (NPT)
(XEN)  - Last Branch Record (LBR) Virtualisation
(XEN)  - Next-RIP Saved on #VMEXIT
(XEN)  - VMCB Clean Bits
(XEN)  - DecodeAssists
(XEN)  - Virtual VMLOAD/VMSAVE
(XEN)  - Virtual GIF
(XEN)  - Pause-Intercept Filter
(XEN)  - Pause-Intercept Filter Threshold
(XEN)  - TSC Rate MSR
(XEN) HVM: SVM enabled
(XEN) HVM: Hardware Assisted Paging (HAP) detected
(XEN) HVM: HAP page sizes: 4kB, 2MB, 1GB
(XEN) Brought up 16 CPUs
(XEN) xenoprof: Initialization failed. AMD processor family 23 is not supported
...
[   1.0000000] NetBSD 9.99.79 (XEN3_DOM0) #31: Fri Feb  5 12:53:17 GMT 2021
...
[   1.0000000] vcpu0: AMD Ryzen 7 1700 Eight-Core Processor          , id 0x800f11
...
[   3.8800164] dk15 at wd6: "efisys", 524288 blocks at 40, type: msdos
[   3.8800164] dk16 at wd6: "root", 4194304 blocks at 524328, type: ffs
...
[   4.0700162] boot device: /dev/dk16
[   4.0700162] unknown device major 0xffffffffffffffff
[   4.0700162] root device: ?
...
[ [  11.6800230] root device: wedge:root
[  14.6900232] dump device: wedge:swap1
[  19.5000230] file system (default generic): 
[  21.5500230] root on dk16 dumps on dk6
[  21.5500230] Your machine does not initialize mem_clusters; sparse_dumps disabled
[  21.5600315] root file system type: ffs
[  21.5600315] kern.module.path=/stand/amd64/9.99.79/modules
[  21.5650467] init path (default /sbin/init): 
...
[  21.5600315] root file system type: ffs
[  21.5600315] kern.module.path=/stand/amd64/9.99.79/modules
[  21.5650467] init path (default /sbin/init): 


Starting xenservices: xenstored, xenconsoled.
Feb  5 14:06:00 quantz xenstored: Checking store ...
Feb  5 14:06:00 quantz xenstored: Checking store complete.
[  50.6500247] balloon0 at xenbus0 id 0: Xen Balloon driver
[  50.6500247] balloon0: current reservation: 4194304 KiB
Setting domain 0 name, domid and JSON config...
Done setting up Dom0
...
Starting cron.
Fri Feb  5 14:06:09 GMT 2021
[  59.9899884] xpq_flush_queue: 1 entries (0 successful) on cpu10 (10)
[  59.9899884] panic: HYPERVISOR_mmu_update failed, ret: -22

[  59.9899884] cpu10: Begin traceback...
[  59.9899884] vpanic() at netbsd:vpanic+0x14a
[  59.9999881] snprintf() at netbsd:snprintf
[  59.9999881] xpq_queue_machphys_update() at netbsd:xpq_queue_machphys_update
[  59.9999881] pmap_install_ptp() at netbsd:pmap_install_ptp+0x198
[  59.9999881] pmap_enter_ma() at netbsd:pmap_enter_ma+0x948
[  59.9999881] pmap_enter() at netbsd:pmap_enter+0x32
[  59.9999881] uvm_fault_internal() at netbsd:uvm_fault_internal+0x15bc
[  59.9999881] trap() at netbsd:trap+0x432
[  59.9999881] --- trap (number 6) ---
[  59.9999881] 7f7ef70041a6:
[  59.9999881] cpu10: End traceback...

[  59.9999881] dumping to dev 168,4 (offset=976762550, size=0): not possible
[  59.9999881] rebooting...
(XEN) Hardware Dom0 shutdown: rebooting machine
(XEN) Resetting with ACPI MEMORY or I/O RESET_REG.

>How-To-Repeat:
Try to boot a xen dom0.
>Fix:

>Audit-Trail:
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@netbsd.org
Cc: port-xen-maintainer@netbsd.org, gnats-admin@netbsd.org,
        netbsd-bugs@netbsd.org
Subject: Re: port-xen/55978: panic: HYPERVISOR_mmu_update failed
Date: Fri, 5 Feb 2021 15:40:37 +0100

 On Fri, Feb 05, 2021 at 02:35:00PM +0000, prlw1@cam.ac.uk wrote:
 > [...]
 > >How-To-Repeat:
 > Try to boot a xen dom0.

 probably not enough, mine boots fine ...

 -- 
 Manuel Bouyer <bouyer@antioche.eu.org>
      NetBSD: 26 ans d'experience feront toujours la difference
 --

From: Patrick Welche <prlw1@cam.ac.uk>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-xen/55978: panic: HYPERVISOR_mmu_update failed
Date: Fri, 5 Feb 2021 14:50:24 +0000

 On Fri, Feb 05, 2021 at 03:40:37PM +0100, Manuel Bouyer wrote:
 > On Fri, Feb 05, 2021 at 02:35:00PM +0000, prlw1@cam.ac.uk wrote:
 > > [...]
 > > >How-To-Repeat:
 > > Try to boot a xen dom0.
 > 
 > probably not enough, mine boots fine ...

 You said that about PR 55888 too, yet you kindly found the bug and fixed it!

 "Try to boot a xen dom0" is what I see: I didn't do anything special
 that I can think of. Just in case:

 # cvs diff -u XEN3_DOM0
 # 

 I included the ryzen bits in the log just in case it's something
 which doesn't happen on intel.

From: Patrick Welche <prlw1@cam.ac.uk>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-xen/55978: panic: HYPERVISOR_mmu_update failed
Date: Sun, 11 Apr 2021 14:29:37 +0100

 On Fri, Feb 05, 2021 at 03:40:37PM +0100, Manuel Bouyer wrote:
 > On Fri, Feb 05, 2021 at 02:35:00PM +0000, prlw1@cam.ac.uk wrote:
 > > [...]
 > > >How-To-Repeat:
 > > Try to boot a xen dom0.
 > 
 > probably not enough, mine boots fine ...

 Since this report, others have reported and reproduced this problem in 
 kern/55667

Home	PR Database Search