Re: [PATCH RFC v2 1/2] crypto: add PKE API

[Tadeusz Struk <tadeusz.struk@intel.com>]

On 05/31/2015 10:48 PM, Herbert Xu wrote:
> On Thu, May 28, 2015 at 09:54:41AM -0700, Tadeusz Struk wrote:
>>
>> If we do this that way then we will be able to pass only one input and one
>> output parameter. There are cases when we will need more that this.
>> For instance for ECDSA signature generation we need one input param hash(m)
>> and two output parameters (r, s).
> 
> There is no reason why you couldn't encode that within one stream.
> As far as as the user is concerned the output is one entity, i.e.,
> the signature.  The fact that it is made up of two numbers is of
> no concern to the API.  It's a technicality for the algorithm to
> sort out.
> 
>> So I have used the SG for that. This is not to deal with non-contiguous memory,
>> but to pass more in/out parameters. Each parameter will need to occupy contiguous space in memory.
>> I will update the comment to make it more clear.
>> If you have other idea how to do this I will be happy to try it.
> 
> If you really wanted to do this then you should be using a simple
> (u8 *, unsigned int) pair but I don't really think this is at all
> necessary.
> 

Ok, I'll rework this to take one char *src and one *dst ptrs.
Thanks
T