Re: [Qemu-devel] [v22 1/2] virtio-crypto: Add virtio crypto device specification

[Halil Pasic <pasic@linux.vnet.ibm.com>]

On 12/11/2017 01:56 PM, Longpeng (Mike) wrote:
> 
> 
> On 2017/12/6 19:01, Halil Pasic wrote:
> 
>>
>>
>> On 12/06/2017 08:37 AM, Longpeng(Mike) wrote:
>>> +\field{outcome_len} is the size of struct virtio_crypto_session_input or
>>> +ZERO for the session-destroy operation.
>>
>> This ain't correct. It should have been something like virtio_crypto_destroy_session_input.
>>
> 
> Hi Halil,
> 
> I already fixed this just now.
> Do you have any other comments on v22 ? I'll send v23 tomorrow if no. :)
> 

Did not read the rest of the document. I'm not in the middle of something,
but I wanted to read the operation part these days. I guess, you prefer
sending out v23 over waiting, so I guess I will wait for v23 then.

Some general questions/remarks before you spin v23:

* I'm not convinced about this 'header' and 'extra parameters' terminology.
Please see https://en.wikipedia.org/wiki/Header_(computing) for header.
I don't think it's fitting for the _flf structs. 
Same about the 'extra'. Please see  https://www.merriam-webster.com/dictionary/extra
(a : more than is due, usual, or necessary : additional) the things in
_vlf aren't extra at all. Do you intend to stick with is terminology?
If yes why? Please explain how should I read/understand it so that it makes
sense!

* Do we want/need to specify any alignment requirement for the
stuff in guest storage (for instance the _flf fields) or be explicit
about no alignment should be assumed (e.g virtio_crypto_mac_create_session_flf.algo
ain't necessarily aligned (in guest memory) as required by uint32_t)?

* I assume one request is supposed to correspond to one descriptor chain.
Right? If yes, could you tell me, where is this expressed in the spec.

Halil