Re: [PATCH] tpm: vtpm_proxy: Do not access host's event log

[Stefan Berger <stefanb-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>]

On 11/18/2016 09:15 AM, Stefan Berger wrote:
> On 11/18/2016 09:11 AM, Stefan Berger wrote:
>> On 11/17/2016 03:37 PM, Jarkko Sakkinen wrote:
>>> On Thu, Nov 17, 2016 at 07:35:05AM -0500, Stefan Berger wrote:
>>>> On 11/16/2016 03:07 PM, Jason Gunthorpe wrote:
>>>>> On Wed, Nov 16, 2016 at 12:07:23PM -0500, Stefan Berger wrote:
>>>>>> The culprit seems to be 'tpm: fix the missing .owner in
>>>>>> tpm_bios_measurements_ops'
>>>>> That is unlikely, it is probably the patch before which calls 
>>>>> read_log
>>>>> unconditionally now. That suggests the crashing is a little random..
>>>> I ran the vtpm driver test suite (with -j32) a few times at that 
>>>> patch and
>>>> it didn't crash. It crashes severely with later patches applied. 
>>>> Here's the
>>>> current experimental patch that fixes these problems:
>>>>
>>>> iff --git a/drivers/char/tpm/tpm_acpi.c b/drivers/char/tpm/tpm_acpi.c
>>>> index 0cb43ef..a73295a 100644
>>>> --- a/drivers/char/tpm/tpm_acpi.c
>>>> +++ b/drivers/char/tpm/tpm_acpi.c
>>>> @@ -56,6 +56,9 @@ int tpm_read_log_acpi(struct tpm_chip *chip)
>>>>
>>>>       log = &chip->log;
>>>>
>>>> +    if (!chip->acpi_dev_handle)
>>>> +        return 0;
>>>> +
>>> If there is a problem in the TPM driver, this does not fix the
>>> problem. It will mask the problem. Maybe there's an ACPI regression
>>> in the rc tree?
>>
>> Following the path from here :
>>
>> http://lxr.free-electrons.com/source/drivers/acpi/acpica/tbxface.c#L282
>>
>> acpi_get_table_with_size -> acpi_tb_validate_table -> 
>> acpi_tb_acquire_table
>>
>> I see acpi_os_map_memory being called in acpi_tb_acquire_table but 
>> not the corresponding acpi_os_unmap_memory...
>>
>
> And to add to this: the call to acpi_get_table() in tpm_acpi.c alone 
> is causing the crash. I am fairly sure that it has something to do 
> with the memory mapping call above not being matched by an unmapping 
> call.

Have to take that all back. This is not the reason. What seems to be the 
reason is just the acpi_os_map_iomem() call. Without calling the 
acpi_os_unmap_iomem() later on, which I assume should be allowed, the 
driver will crash when the device terminates.

     Stefan

>
>
>>     Stefan
>>
>>
>>>
>>> This is a funky situation because those lines need to be there but
>>> I do not want them before it is root caused that it is not a TPM
>>> bug.
>>>
>>> /Jarkko
>>>
>>
>


------------------------------------------------------------------------------