OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

virtio-dev message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [Qemu-devel] [PATCH v16 1/2] virtio-crypto: Add virtio crypto device specification



On 02/13/2017 02:41 PM, Stefan Hajnoczi wrote:
> On Fri, Feb 03, 2017 at 07:07:02PM +0100, Halil Pasic wrote:
>>
>>
>> On 02/03/2017 04:37 PM, Michael S. Tsirkin wrote:
>>> On Fri, Feb 03, 2017 at 10:33:16AM +0000, Stefan Hajnoczi wrote:
>>>>> +\begin{description}
>>>>> +\item[0] dataq1
>>>>> +\item[\ldots]
>>>>> +\item[N-1] dataqN
>>>>> +\item[N] controlq
>>>>> +\end{description}
>>>>> +
>>>>> +N is set by \field{max_dataqueues}.
>>>>> +
>>>>> +\subsection{Feature bits}\label{sec:Device Types / Crypto Device / Feature bits}
>>>>> +
>>>>> +VIRTIO_CRYPTO_F_NON_SESSION_MODE (0) non-session mode is available.
>>>>> +VIRTIO_CRYPTO_F_CIPHER_NON_SESSION_MODE (1) non-session mode is available for CIPHER service.
>>>>> +VIRTIO_CRYPTO_F_HASH_NON_SESSION_MODE (2) non-session mode is available for HASH service.
>>>>> +VIRTIO_CRYPTO_F_MAC_NON_SESSION_MODE (3) non-session mode is available for MAC service.
>>>>> +VIRTIO_CRYPTO_F_AEAD_NON_SESSION_MODE (4) non-session mode is available for AEAD service.
>>>>
>>>> Bits 1-4 require bit 0.  Is bit 0 necessary at all?  Or may bits 1-4 can
>>>> be eliminated in favor of just bit 0.
>>>
>>> Too late to change it as QEMU released a device with that feature.
                            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>>>
>>
>> When was this introduced to QEMU? Can't find it in current master
>> and I do not remember seeing the patches.
> 
> s/QEMU/Linux/
> 
> It's in drivers/crypto/virtio/.
> 


Thanks! Sorry but I could not find the NON_SESSION_MODE in Linus master
(Linux) either, and Michael has stated that a QEMU device was released
with these. I'm still confused. But if it's only Linux, it most likely is
not interface, so theoretically should not break anything if we change it
-- should we want to change it.

I think your original questions regarding these feature bits are valid
(and have not been addressed in a satisfactory way yet) by the way.

Halil





Attachment: signature.asc
Description: OpenPGP digital signature



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]