On 18.11.21 03:37, Stefano Stabellini wrote:
> On Wed, 17 Nov 2021, Jan Beulich wrote:
>> On 17.11.2021 03:11, Stefano Stabellini wrote:
>>> --- a/drivers/xen/xenbus/xenbus_probe.c
>>> +++ b/drivers/xen/xenbus/xenbus_probe.c
>>> @@ -951,6 +951,18 @@ static int __init xenbus_init(void)
>>>   		err = hvm_get_parameter(HVM_PARAM_STORE_PFN, &v);
>>>   		if (err)
>>>   			goto out_error;
>>> +		/*
>>> +		 * Uninitialized hvm_params are zero and return no error.
>>> +		 * Although it is theoretically possible to have
>>> +		 * HVM_PARAM_STORE_PFN set to zero on purpose, in reality it is
>>> +		 * not zero when valid. If zero, it means that Xenstore hasn't
>>> +		 * been properly initialized. Instead of attempting to map a
>>> +		 * wrong guest physical address return error.
>>> +		 */
>>> +		if (v == 0) {
>>> +			err = -ENOENT;
>>> +			goto out_error;
>>> +		}
>>
>> If such a check gets added, then I think known-invalid frame numbers
>> should be covered at even higher a priority than zero.
> 
> Uhm, that's a good point. We could check for 0 and also ULONG_MAX
> 
> 
>> This would, for example, also mean to ...
>>
>>>   		xen_store_gfn = (unsigned long)v;
>>
>> ... stop silently truncating a value here.
> 
> Yeah, it can only happen on 32-bit but you have a point.
> 
> 
>> By covering them we would then have the option to pre-fill PFN params
>> with, say, ~0 in the hypervisor (to clearly identify them as invalid,
>> rather than having to guess at the validity of 0). I haven't really
>> checked yet whether such a change would be compatible with existing
>> software ...
> 
> I had the same idea. I think the hvm_params should be initialized to an
> invalid value in Xen. But here in Linux we need to be able to cope with
> older Xen versions too so it still makes sense to check for zero in
> places where it is very obviously incorrect (HVM_PARAM_STORE_PFN).
> 
> 
> What do you think of the appended?
> 
> 
> 
> diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c
> index 94405bb3829e..04558d3a5562 100644
> --- a/drivers/xen/xenbus/xenbus_probe.c
> +++ b/drivers/xen/xenbus/xenbus_probe.c
> @@ -951,6 +951,28 @@ static int __init xenbus_init(void)
>   		err = hvm_get_parameter(HVM_PARAM_STORE_PFN, &v);
>   		if (err)
>   			goto out_error;
> +		/*
> +		 * Return error on an invalid value.
> +		 *
> +		 * Uninitialized hvm_params are zero and return no error.
> +		 * Although it is theoretically possible to have
> +		 * HVM_PARAM_STORE_PFN set to zero on purpose, in reality it is
> +		 * not zero when valid. If zero, it means that Xenstore hasn't
> +		 * been properly initialized. Instead of attempting to map a
> +		 * wrong guest physical address return error.
> +		 */
> +		if (v == 0) {

Make this "if (v == ULONG_MAX || v== 0)" instead?
This would result in the same err on a new and an old hypervisor
(assuming we switch the hypervisor to init params with ~0UL).

> +			err = -ENOENT;
> +			goto out_error;
> +		}
> +		/*
> +		 * ULONG_MAX is invalid on 64-bit because is INVALID_PFN.
> +		 * On 32-bit return error to avoid truncation.
> +		 */
> +		if (v >= ULONG_MAX) {
> +			err = -EINVAL;
> +			goto out_error;
> +		}

Does it make sense to continue the system running in case of
truncation? This would be a 32-bit guest with more than 16TB of RAM
and the Xen tools decided to place the Xenstore ring page above the
16TB boundary. This is a completely insane scenario IMO.

A proper panic() in this case would make diagnosis of that much
easier (me having doubts that this will ever be hit, though).


Juergen