Re: kernel 4.19.5 crashes in Xen DomU
On Mon, 3 Dec 2018, Markus Schönhaber wrote:
Hi,
I'm running Fedora 28 in a Xen DomU on a Debian Stretch Dom0. This used
to work fine but kernel 4.19.5 crashes on boot (see console output
below) - 4.19.4 and earlier don't show any problems.
Is anyone else seeing this? Is there a workaround?
Regards
mks
...
>
> [ 1.919053] BUG: unable to handle kernel paging request at ffff88884b969000
> [ 1.919061] PGD 220c067 P4D 220c067 PUD 0
> [ 1.919067] Oops: 0000 [#1] SMP NOPTI
> [ 1.919072] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.19.5-200.fc28.x86_64 #1
> [ 1.919080] RIP: e030:ptdump_walk_pgd_level_core+0x21a/0x4f0
> [ 1.919085] Code: b8 00 00 00 00 00 40 00 00 49 83 c7 08 49 01 c5 4c 3b 7c 24 40
0f 84 c2 02 00 00 4c 89 e8 48 c1 f8 10 48 89 84 24 80 00 00 00 <49> 8b 3f 48 f7 c7
9f ff ff ff 0f 85 22 ff ff ff 41 b8 03 00 00 00
> [ 1.919094] RSP: e02b:ffffc90000717e68 EFLAGS: 00010286
> [ 1.919099] RAX: ffff800000000000 RBX: ffffffff8191c029 RCX: 0000000000000000
> [ 1.919104] RDX: 0000000000000101 RSI: ffffffff8220a000 RDI: 000000084b969067
> [ 1.919109] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000
> [ 1.919114] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
> [ 1.919118] R13: 8000000000000000 R14: 0000000000000000 R15: ffff88884b969000
> [ 1.919128] FS: 0000000000000000(0000) GS:ffff88807d400000(0000)
knlGS:0000000000000000
> [ 1.919133] CS: e033 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 1.919137] CR2: ffff88884b969000 CR3: 000000000220a000 CR4: 0000000000042660
> [ 1.919144] Call Trace:
> [ 1.919151] ? printk+0x58/0x6f
> [ 1.919157] ? rest_init+0xaa/0xaa
> [ 1.919161] kernel_init+0x31/0x107
> [ 1.919166] ret_from_fork+0x35/0x40
> [ 1.919170] Modules linked in:
> [ 1.919175] CR2: ffff88884b969000
> [ 1.919179] ---[ end trace a85adfe208024c43 ]---
> [ 1.919185] RIP: e030:ptdump_walk_pgd_level_core+0x21a/0x4f0
> [ 1.919190] Code: b8 00 00 00 00 00 40 00 00 49 83 c7 08 49 01 c5 4c 3b 7c 24 40
0f 84 c2 02 00 00 4c 89 e8 48 c1 f8 10 48 89 84 24 80 00 00 00 <49> 8b 3f 48 f7 c7
9f ff ff ff 0f 85 22 ff ff ff 41 b8 03 00 00 00
> [ 1.919199] RSP: e02b:ffffc90000717e68 EFLAGS: 00010286
> [ 1.919204] RAX: ffff800000000000 RBX: ffffffff8191c029 RCX: 0000000000000000
> [ 1.919209] RDX: 0000000000000101 RSI: ffffffff8220a000 RDI: 000000084b969067
> [ 1.919214] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000
> [ 1.919218] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
> [ 1.919223] R13: 8000000000000000 R14: 0000000000000000 R15: ffff88884b969000
> [ 1.919231] FS: 0000000000000000(0000) GS:ffff88807d400000(0000)
knlGS:0000000000000000
> [ 1.919237] CS: e033 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 1.919241] CR2: ffff88884b969000 CR3: 000000000220a000 CR4: 0000000000042660
> [ 1.919261] Kernel panic - not syncing: Attempted to kill init!
exitcode=0x00000009
> [ 1.919261]
> [ 1.919269] Kernel Offset: disabled
It looks like the "Linux 4.19.5 fails to boot as Xen dom0" issue being
discussed on the xen-devel list, eg. see
https://lists.xenproject.org/archives/html/xen-devel/2018-11/msg03557.html
Basically there is an issue with the 4.19.5 kernel that upset Xen.
Michael Young