Re: [Xen-devel] ACPI / ioremap() crash

[Juergen Gross <jgross@xxxxxxxx>]

On 26/02/18 10:05, Jan Beulich wrote:
> Boris, Jürgen,
> 
> now for the actual crash:
> 
> ACPI: Core revision 20170831
> BUG: unable to handle kernel paging request at ffff8801d8c09050
> IP: xen_set_pmd+0x3a/0x50
> PGD 1c0a067 P4D 1c0a067 PUD 1de2067 PMD 1d9b3d067 PTE 80100001d8c09065
> Oops: 0003 [#1] SMP
> Modules linked in:
> Supported: Yes
> CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.15.3-2018-02-12-xen0 #1
> Hardware name: ...
> RIP: e030:xen_set_pmd+0x3a/0x50
> RSP: e02b:ffffffff81c03bc0 EFLAGS: 00010246
> RAX: 0200000000000800 RBX: ffff880187c2e000 RCX: 0000000000000080
> RDX: ffffea0000000000 RSI: 0000000224f4f067 RDI: ffff8801d8c09050
> RBP: ffff8801d8c09050 R08: ffffea00055b2a10 R09: 00000000055b2a48
> R10: 6db6db6db6db6db7 R11: ffff880000000000 R12: ffffffff81c38280
> R13: 0000000000187c2e R14: ffff880207c2e000 R15: ffff8801d8c09050
> FS:  0000000000000000(0000) GS:ffff880259e00000(0000) knlGS:0000000000000000
> CS:  e033 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: ffff8801d8c09050 CR3: 0000000001c09000 CR4: 0000000000002660
> Call Trace:
>  ? __pte_alloc_kernel+0xba/0x100
>  ? __raw_callee_save_xen_pud_val+0x11/0x1e
>  ? ioremap_page_range+0x2e8/0x3f0
>  ? __ioremap_caller+0x1dd/0x300
>  ? acpi_os_map_iomem+0x135/0x1e0
>  ? acpi_os_map_iomem+0x135/0x1e0
>  ? acpi_tb_acquire_table+0x38/0x66
>  ? acpi_tb_validate_table+0x1e/0x31
>  ? acpi_tb_verify_temp_table+0x1d/0x1f3
>  ? acpi_reallocate_root_table+0xa6/0xe4
>  ? dmi_check_system+0x19/0x50
>  ? acpi_early_init+0x4f/0x118
>  ? start_kernel+0x296/0x39b
>  ? set_init_arg+0x5f/0x5f
>  ? xen_start_kernel+0x34d/0x569
> Code: 48 01 fa 48 0f 42 05 06 97 bf 00 48 01 d0 48 ba 00 00 00 00 00 ea ff ff 
> 48 c1 e8 0c 48 6b c0 38 48 01 d0 48 8b 00 f6 c4 02 75 04 <48> 89 37 c3 e9 2d 
> ff ff ff 0f 1f 00 66 2e 0f 1f 84 00 00 00 00 
> RIP: xen_set_pmd+0x3a/0x50 RSP: ffffffff81c03bc0
> CR2: ffff8801d8c09050
> ---[ end trace 7e64b9529af2b368 ]---
> Kernel panic - not syncing: Attempted to kill the idle task!
> 
> Clearly the kernel tries to write to a r/o L2 table here. I suppose
> the system somehow managed to run into an L2 table that hasn't
> been marked as pinned. Do you have any immediate idea, before
> I can find time to actually debug this?
> 
> The issue didn't exist with 4.14.1, if that helps / matters.

Upstream commit 895f7b8e90200cf1a5dc313329369adf30e51f9a addresses
this issue (contained in 4.16-rc3, flagged for stable).


Juergen

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel