List moderator: feel free to delete my previous large message with attachments that's in the moderation queue...it's now obsolete anyway.


I have found a fix/workaround for my reboot issues with Xen 4.6.3-12 + Kernel 4.9.13:

Once I finally got serial output all the way through the boot process (xen+dom0) I discovered the stack trace:

[Firmware Bug]: CPU7: APIC id mismatch. Firmware: 0 APIC: 7
installing Xen timer for CPU 8
[Firmware Bug]: CPU8: APIC id mismatch. Firmware: 0 APIC: 20
smpboot: Package 1 of CPU 8 exceeds BIOS package data 1.
------------[ cut here ]------------
kernel BUG at arch/x86/kernel/cpu/common.c:997!
invalid opcode: 0000 [#1] SMP
Modules linked in:
CPU: 8 PID: 0 Comm: swapper/8 Not tainted 4.9.13-22.el7.x86_64 #1
Hardware name: Supermicro X9DRT/X9DRT, BIOS 3.2a 08/04/2015
random: fast init done
task: ffff880058a8c4c0 task.stack: ffffc900400b4000
RIP: e030:[<ffffffff8103e527>]  [<ffffffff8103e527>] identify_secondary_cpu+0x57/0x80
RSP: e02b:ffffc900400b7f08  EFLAGS: 00010086
RAX: 00000000ffffffe4 RBX: ffff88005d80a020 RCX: ffffffff81c5be68
RDX: 0000000000000001 RSI: 0000000000000005 RDI: 0000000000000005
RBP: ffffc900400b7f18 R08: 00000000000000cb R09: 0000000000000004
R10: 0000000000000000 R11: 0000000000000006 R12: 0000000000000008
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
FS:  0000000000000000(0000) GS:ffff88005d800000(0000) knlGS:0000000000000000
CS:  e033 DS: 002b ES: 002b CR0: 0000000080050033
CR2: 0000000000000000 CR3: 0000000001c07000 CR4: 0000000000042660
Stack:
 0000000000000008 0000000000000000 ffffc900400b7f28 ffffffff8104e94e
 ffffc900400b7f40 ffffffff81029925 0000000000000000 ffffc900400b7f50
 ffffffff810299a0 0000000000000000 0000000000000000 0000000000000000
Call Trace:
 [<ffffffff8104e94e>] smp_store_cpu_info+0x3e/0x40
 [<ffffffff81029925>] cpu_bringup+0x35/0x90
 [<ffffffff810299a0>] cpu_bringup_and_idle+0x20/0x40
Code: 44 89 e7 ff 50 68 0f b7 93 d2 00 00 00 39 d0 75 1c 0f b7 bb da 00 00 00 44 89 e6 e8 24 03 01 00 85 c0 75 07 5b 41 5c 5d c3 0f 0b <0f> 0b 0f b7 8b d4 00 00 00 89 c2 44 89 e6 48 c7 c7 98 87 a6 81
RIP  [<ffffffff8103e527>] identify_secondary_cpu+0x57/0x80
 RSP <ffffc900400b7f08>
---[ end trace dc5563100443876e ]---

I surmised that reducing the number of dom0 vcpu might solve this issue (they were unbounded)

In testing adding "dom0_max_vcpus=4 dom0_vcpus_pin" to the GRUB_CMDLINE_XEN_DEFAULT line in /etc/defaults/grub and re-running grub2-mkconfig has resulted in the system I have that never booted Xen 4.6.3-12 + Kernel 4.9.13, booting every single time out of 5-10 tests.


So...I don't know if there's a race condition somewhere, or what...but...so far this workaround has not failed me.

Thanks,
-Dave



> On Fri, Apr 7, 2017 at 6:58 AM, PJ Welsh <pjwelsh at gmail.com
>> wrote:
>> I've not gotten any bites from my posting on the xen-devel mailing list.
>> Here is the only one to-date:
>> https://lists.xen.org/archives/html/xen-devel/2017-04/msg01069.html
>>
>> From that email, there needs to be some hypervisor messages.
>>
>> Does anyone know how to produce the hypervisor messages? I've already
>
>> removed the rhgb and quiet options from the boot.
>
>>
>> Thanks
>> PJ
>
>
> I spoke too soon. To get more information: Please see
>
> https://wiki.xenproject.org/wiki/Reporting_Bugs_against_Xen_Project
>
> and
>
> https://wiki.xenproject.org/wiki/Xen_Serial_Console
>
> or alternatively at least add "vga=keep".
>
> pjwelsh

_______________________________________________
CentOS-virt mailing list
CentOS-virt@centos.org
https://lists.centos.org/mailman/listinfo/centos-virt