[linux-sunxi] Re: [PATCH v9 4/4] crypto: Add Allwinner Security System crypto accelerator

[Corentin LABBE]

Le 24/05/2015 05:32, Herbert Xu a écrit :
> On Sat, May 23, 2015 at 04:35:36PM +0200, Boris Brezillon wrote:
>>
>>> Since all SG I get was always a multiple of 16 (AES BLOCK SIZE) it was a sort of confirmation.
>>>
>>> Herbert ? does am I right or a chunking test is missing for cbc(aes) in testmgr.h
>>
>> Okay, just sharing my vision of this thing (I'll let Herbert comment on
>> this aspect): I'd say that theoretically nothing prevents one from
>> splitting its sg list in chunks smaller than the block size, so I'd
>> say you should use the same trick for AES. 
> 
> Indeed, you must be able to handle a block that straddles two SG
> entries.  If the hardware is not capable then you have to linearise
> it by copying or use a fallback.
> 
> I can't believe we don't have a generic test for this in testmgr.
> 

For aes_cbc it exists a test with 3 SG with .tap = { 496 - 20, 4, 16 }
But my driver handle that. (multiple of 4)

What do you think about adding a test with 16 SG of 1 byte ? (or 3 + 2 + 3 + 8 * 1)

Furthermore, I need to re-read testmgr.h but it seems that no aes test exists with odd SG size.