Re: [Xen-devel] [PATCH v3] xen: fix reboot/shutdown with running HVM guests

[Aravind Gopalakrishnan <aravind.gopalakrishnan@xxxxxxx>]

On 6/5/2014 10:01 AM, Roger Pau Monne wrote:
> If there's a guest using VMX/SVM when the hypervisor shuts down, it
> can lead to the following crash due to VMX/SVM functions being called
> after hvm_cpu_down has been called. In order to prevent that, check in
> {svm/vmx}_ctxt_switch_from that the cpu virtualization extensions are
> still enabled.
>
> (XEN) Domain 0 shutdown: rebooting machine.
> (XEN) Assertion 'read_cr0() & X86_CR0_TS' failed at vmx.c:644
> (XEN) ----[ Xen-4.5-unstable  x86_64  debug=y  Tainted:    C ]----
> (XEN) CPU:    0
> (XEN) RIP:    e008:[<ffff82d0801d90ce>] vmx_ctxt_switch_from+0x1e/0x14c
> (XEN) RFLAGS: 0000000000010046   CONTEXT: hypervisor
> (XEN) rax: 0000000080050033   rbx: ffff8300dfb1c000   rcx: 0000000000000000
> (XEN) rdx: 0000000000000000   rsi: ffff82d0802d7fc0   rdi: ffff8300dfb1c000
> (XEN) rbp: ffff82d0802d7a88   rsp: ffff82d0802d7a78   r8:  0000000000000000
> (XEN) r9:  ffff82cffffff000   r10: 0000000b06dca869   r11: 0000003d7d708160
> (XEN) r12: ffff8300dfb1c000   r13: 0000000000000000   r14: ffff82d0802d0000
> (XEN) r15: ffff82d0802d7f18   cr0: 0000000080050033   cr4: 00000000000026f0
> (XEN) cr3: 000000019ed8d000   cr2: 0000000800dcb040
> (XEN) ds: 0000   es: 0000   fs: 0000   gs: 0000   ss: 0000   cs: e008
> (XEN) Xen stack trace from rsp=ffff82d0802d7a78:
> (XEN)    ffff8300dfdf2000 ffff8300dfdf2000 ffff82d0802d7ad8 ffff82d08015d129
> (XEN)    fffffe003d272a38 ffff83019a3f9140 0000000000000000 0000000000000001
> (XEN)    0000000000000086 000000000019a400 0000000000000000 0001000000010030
> (XEN)    ffff82d0802d7af8 ffff82d080160acf ffff83019ec18530 ffff8300dfdf2000
> (XEN)    ffff82d0802d7b08 ffff82d080160af9 ffff82d0802d7b58 ffff82d080161728
> (XEN)    ffff82d0802d7b48 ffff82d08013ffbf 0000000000000002 ffff83019ec18530
> (XEN)    0000000000000000 0000000000000012 0000000000000000 0001000000010030
> (XEN)    ffff82d0802d7b68 ffff82d08014e721 ffff82d0802d7bc8 ffff82d08014cda2
> (XEN)    ffff8300dfb1c000 0000000000000092 ffff83019ec18604 ffff83019ec185f8
> (XEN)    ffff82d0802d0000 0000000000000001 0000000000000000 ffff82d08016560e
> (XEN)    ffff82d080272860 0000000000000020 ffff82d0802d7bd8 ffff82d0801448a8
> (XEN)    ffff82d0802d7be8 ffff82d080165625 ffff82d0802d7c18 ffff82d080166143
> (XEN)    0000000000000000 0000000000000001 0000000000000000 ffff82d080272860
> (XEN)    ffff82d0802d7c48 ffff82d080166aa8 ffff8300dfb1c060 0000000000010000
> (XEN)    0000000000000001 ffff82d080272860 ffff82d0802d7c78 ffff82d080166bae
> (XEN)    000000000000000a ffff82d080276fe0 00000000000000fb 0000000000000061
> (XEN)    ffff82d0802d7c98 ffff82d080166f63 ffff82d0802d7c98 ffff82d0801821ff
> (XEN)    ffff82d0802d7cb8 ffff82d08018228b 0000000000000000 ffff82d0802d7dd8
> (XEN)    ffff82d0802d7cf8 ffff82d080181aa7 ffff82d0802d7d08 0000000000000206
> (XEN)    0000000000000000 ffff82d0802d7dd8 00000000000000fb 0000000000000008
> (XEN) Xen call trace:
> (XEN)    [<ffff82d0801d90ce>] vmx_ctxt_switch_from+0x1e/0x14c
> (XEN)    [<ffff82d08015d129>] __context_switch+0x127/0x462
> (XEN)    [<ffff82d080160acf>] __sync_local_execstate+0x6a/0x8b
> (XEN)    [<ffff82d080160af9>] sync_local_execstate+0x9/0xb
> (XEN)    [<ffff82d080161728>] map_domain_page+0x88/0x4de
> (XEN)    [<ffff82d08014e721>] map_vtd_domain_page+0xd/0xf
> (XEN)    [<ffff82d08014cda2>] io_apic_read_remap_rte+0x158/0x29f
> (XEN)    [<ffff82d0801448a8>] iommu_read_apic_from_ire+0x27/0x29
> (XEN)    [<ffff82d080165625>] io_apic_read+0x17/0x65
> (XEN)    [<ffff82d080166143>] __ioapic_read_entry+0x38/0x61
> (XEN)    [<ffff82d080166aa8>] clear_IO_APIC_pin+0x1a/0xf3
> (XEN)    [<ffff82d080166bae>] clear_IO_APIC+0x2d/0x60
> (XEN)    [<ffff82d080166f63>] disable_IO_APIC+0xd/0x81
> (XEN)    [<ffff82d08018228b>] smp_send_stop+0x58/0x68
> (XEN)    [<ffff82d080181aa7>] machine_restart+0x80/0x20a
> (XEN)    [<ffff82d080181c3c>] __machine_restart+0xb/0xf
> (XEN)    [<ffff82d080128fb9>] smp_call_function_interrupt+0x99/0xc0
> (XEN)    [<ffff82d080182330>] call_function_interrupt+0x33/0x43
> (XEN)    [<ffff82d08016bd89>] do_IRQ+0x9e/0x63a
> (XEN)    [<ffff82d08016406f>] common_interrupt+0x5f/0x70
> (XEN)    [<ffff82d0801a8600>] mwait_idle+0x29c/0x2f7
> (XEN)    [<ffff82d08015cf67>] idle_loop+0x58/0x76
> (XEN)
> (XEN)
> (XEN) ****************************************
> (XEN) Panic on CPU 0:
> (XEN) Assertion 'read_cr0() & X86_CR0_TS' failed at vmx.c:644
> (XEN) ****************************************
> (XEN)
> (XEN) Reboot in five seconds...
>
> Signed-off-by: Roger Pau Monnà <roger.pau@xxxxxxxxxx>
> Suggested-by: Jan Beulich <jbeulich@xxxxxxxx>
> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>
> Cc: Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>
> Cc: Suravee Suthikulpanit <suravee.suthikulpanit@xxxxxxx>
> Cc: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@xxxxxxx>
> Cc: Keir Fraser <keir@xxxxxxx>
> Cc: Jan Beulich <jbeulich@xxxxxxxx>
> Cc: Jun Nakajima <jun.nakajima@xxxxxxxxx>
> Cc: Eddie Dong <eddie.dong@xxxxxxxxx>
> Cc: Kevin Tian <kevin.tian@xxxxxxxxx>
> Cc: Mukesh Rathor <mukesh.rathor@xxxxxxxxxx>
> ---
> Changes since v2:
>   - Added comments on the gates added to {svm/vmx}_ctxt_switch_from.
>   - Move the declaration of the pcpu variable vmxon to vmcs.h.
> ---
>   xen/arch/x86/hvm/svm/svm.c         |    8 ++++++++
>   xen/arch/x86/hvm/vmx/vmcs.c        |    2 +-
>   xen/arch/x86/hvm/vmx/vmx.c         |    8 ++++++++
>   xen/include/asm-x86/hvm/vmx/vmcs.h |    2 ++
>   4 files changed, 19 insertions(+), 1 deletions(-)
Acked-by: Aravind Gopalakrishnan <Aravind.Gopalakrishnan@xxxxxxx>


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel