Re: [PATCH v1 3/4] OvmfPkg: IommuDxe: Zero the shared page(s) on Unmap()

[Laszlo Ersek <lersek@redhat.com>]

On 08/02/17 13:22, Brijesh Singh wrote:
> 
> 
> On 8/2/17 2:37 AM, Laszlo Ersek wrote:
>> //
>>> +  // Zero the shared memory so that hypervisor no longer able to get intelligentable
>>> +  // data.
>>> +  //
>>> +  SetMem ((VOID *) (UINTN)MapInfo->DeviceAddress, MapInfo->NumberOfBytes, 0);
>> Please use ZeroMem().
>>
>> Furthermore, ZeroMem() should occur just before every FreePages() call:
>> - when Unmap() releases the implicitly allocated bounce buffer
>> - when FreeBuffer() releases the explicitly allocated common buffer
>>   (I thought I spelled this out in my previous email(s), but in
>>   retrospect it seems I only intended to :/ )
>> - in the virtio drivers' exit-boot-services callbacks, FreeBuffer()
>>   can't be called (only Unmap(), after the virtio reset), so the
>>   ZeroMem() should be done manually there.
> 
> Not sure why do we need to ZeroMem() when FreeBuffer()  is called for
> explicitly allocated common buffer ? I thought before calling the
> FreeBuffer() on common buffer, client will call Unmap() which will
> restore the C-bit state on the common buffer and also update the
> contents (i.e now common buffer will contain encrypted data).
> 

My bad, you are totally right -- when I wrote the above, I actually
reviewed the "BusMasterCommonBuffer" section of my earlier message

http://mid.mail-archive.com/e1137140-f5d5-7eb8-0162-952694b24f96@redhat.com

and I totally missed that in that message I had written

"Client calls Unmap(). Unmap() restores the C bit in one fell swoop, and
encrypts the buffer in-place (by bouncing it page-wise to the static
array and back)."

Sigh. Need more rest.

Thanks for catching my error!
Laszlo