|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] [PATCH 1/2] x86/vmx: Print the problematic MSR if a vmentry fails
On 13/10/16 12:38, Jan Beulich wrote:
>>>> On 13.10.16 at 13:15, <andrew.cooper3@xxxxxxxxxx> wrote:
>> Sample error looks like:
>>
>> (XEN) Failed vm entry (exit reason 0x80000022) caused by MSR loading
>> (entry 13).
>> (XEN) msr 0000068a, val 1fff800000102af0, (mbz 0)
>> (XEN) ************* VMCS Area **************
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
> with
>
>> --- a/xen/arch/x86/hvm/vmx/vmx.c
>> +++ b/xen/arch/x86/hvm/vmx/vmx.c
>> @@ -3104,8 +3104,23 @@ static void vmx_failed_vmentry(unsigned int
>> exit_reason,
>> printk("caused by invalid guest state (%ld).\n",
>> exit_qualification);
>> break;
>> case EXIT_REASON_MSR_LOADING:
>> - printk("caused by MSR entry %ld loading.\n", exit_qualification);
>> + {
>> + unsigned long idx = exit_qualification - 1;
>> + struct vmx_msr_entry *msr;
> ... preferably const here
Will do.
> (and the declaration perhaps moved into
> the more narrow scope it's needed in), ...
Cant, due to its use in the sizeof() expression. (I tried that first)
>
>> + printk("caused by MSR loading (entry %ld).\n", idx);
> ... %lu (plus ideally the full stop dropped) here, and ...
Ok for %lu. The full stop is dealt with in the following patch, so will
leave this as-is for consistency of both patches.
>
>> + if ( idx > (PAGE_SIZE / sizeof(*msr)) )
> ... >= here.
>
>> + printk(" Entry out of range\n");
>> + else
>> + {
>> + msr = &curr->arch.hvm_vmx.msr_area[idx];
>> +
>> + printk(" msr %08x, val %016"PRIx64", (mbz %#x)\n",
> I'm also unsure about the usefulness of the commas here - at least
> the second one seems a little strange with the value in parentheses.
Ok.
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel
|
![]() |
Lists.xenproject.org is hosted with RackSpace, monitoring our |