Re: [PATCH 5/7] KVM: TDX: Handle TDG.VP.VMCALL<ReportFatalError>

[Binbin Wu <binbin.wu@linux.intel.com>]

On 12/10/2024 5:05 PM, Chao Gao wrote:
>> diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
>> index edc070c6e19b..bb39da72c647 100644
>> --- a/Documentation/virt/kvm/api.rst
>> +++ b/Documentation/virt/kvm/api.rst
>> @@ -6815,6 +6815,7 @@ should put the acknowledged interrupt vector into the 'epr' field.
>>    #define KVM_SYSTEM_EVENT_WAKEUP         4
>>    #define KVM_SYSTEM_EVENT_SUSPEND        5
>>    #define KVM_SYSTEM_EVENT_SEV_TERM       6
>> +  #define KVM_SYSTEM_EVENT_TDX_FATAL      7
>> 			__u32 type;
>>                          __u32 ndata;
>>                          __u64 data[16];
>> @@ -6841,6 +6842,13 @@ Valid values for 'type' are:
>>     reset/shutdown of the VM.
>>   - KVM_SYSTEM_EVENT_SEV_TERM -- an AMD SEV guest requested termination.
>>     The guest physical address of the guest's GHCB is stored in `data[0]`.
>> + - KVM_SYSTEM_EVENT_TDX_FATAL -- an TDX guest requested termination.
> Not sure termination is an accurate interpretation of fatal errors. Maybe
> just say: a fatal error reported by a TDX guest.
OK, will update it as:
"a TDX guest reported a fatal error state."

>
>> +   The error codes of the guest's GHCI is stored in `data[0]`.
> what do you mean by "guest's GHCI"?
I don't know what I was thinking about.

Will update it as:
    The error code reported by the TDX guest is stored in `data[0]`, the error
    code format is defined in TDX GHCI specification.

>
>> +   If the bit 63 of `data[0]` is set, it indicates there is TD specified
>> +   additional information provided in a page, which is shared memory. The
>> +   guest physical address of the information page is stored in `data[1]`.
>> +   An optional error message is provided by `data[2]` ~ `data[9]`, which is
>> +   byte sequence, LSB filled first. Typically, ASCII code(0x20-0x7e) is filled.
>>   - KVM_SYSTEM_EVENT_WAKEUP -- the exiting vCPU is in a suspended state and
>>     KVM has recognized a wakeup event. Userspace may honor this event by
>>     marking the exiting vCPU as runnable, or deny it and call KVM_RUN again.