RE: [EXT] [PATCH v6 3/4] crypto: caam - add in-kernel interface for blob generator

[Pankaj Gupta]

Hi Ahmad,

Please find the comments in-line.

Regards
Pankaj

> -----Original Message-----
> From: Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx>
> Sent: Thursday, March 24, 2022 3:40 PM
> To: Pankaj Gupta <pankaj.gupta@xxxxxxx>; Horia Geanta
> <horia.geanta@xxxxxxx>; Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>;
> David S. Miller <davem@xxxxxxxxxxxxx>
> Cc: linux-security-module@xxxxxxxxxxxxxxx; Eric Biggers
> <ebiggers@xxxxxxxxxx>; David Gstir <david@xxxxxxxxxxxxx>; Matthias
> Schiffer <matthias.schiffer@xxxxxxxxxxxxxxx>; Sumit Garg
> <sumit.garg@xxxxxxxxxx>; Jan Luebbe <j.luebbe@xxxxxxxxxxxxxx>; Richard
> Weinberger <richard@xxxxxx>; tharvey@xxxxxxxxxxxxx; Franck Lenormand
> <franck.lenormand@xxxxxxx>; James Morris <jmorris@xxxxxxxxx>; Mimi
> Zohar <zohar@xxxxxxxxxxxxx>; linux-kernel@xxxxxxxxxxxxxxx; David Howells
> <dhowells@xxxxxxxxxx>; Jarkko Sakkinen <jarkko@xxxxxxxxxx>;
> keyrings@xxxxxxxxxxxxxxx; linux-crypto@xxxxxxxxxxxxxxx;
> kernel@xxxxxxxxxxxxxx; linux-integrity@xxxxxxxxxxxxxxx; James Bottomley
> <jejb@xxxxxxxxxxxxx>; Serge E. Hallyn <serge@xxxxxxxxxx>
> Subject: Re: [EXT] [PATCH v6 3/4] crypto: caam - add in-kernel interface for
> blob generator
> 
> Caution: EXT Email
> 
> Hello Pankaj,
> 
> On 24.03.22 10:55, Pankaj Gupta wrote:
> > Hi Ahmad,
> >
> > Please find the comments in-line.
> 
> Thanks for you review.
> 
> > Suggest to continue to use two separate descriptor-creation-function for
> 'encap' and 'decap'.
> > This will help these API(s) to be maintained easily going forward.
> 
> We can still split them up in future once there is a real need.
> But currently they are exactly the same, except for input/output length, so I
> think it's correct to not introduce duplication unless needed.
> 
> >>   - use append_seq_(in|out)_ptr_intlen for both encap/decap as a
> >> result
> 
> Case in point. The intlen omission was because the two functions are largely
> identical and I only fixed up one of them. This is prone to repeat when we go
> back to have identical code with minor differences.
> 
> > In continuation to the previous comment, there is another suggestion:
> >
> > Either:
> > struct keyblob_info {
> >         void *key;
> >         size_t key_len;
> >
> >         void *blob;
> >         size_t blob_len;
> >
> >         size_t key_mod_len;
> >         const void *key_mod;
> > };
> 
> I can do that.
> 

Please do. Thanks.
Patch 4/4, needs to be re-worked to.

> Whats your opinion on the desc size computation? Comment the macro or
> add the static inline helper?
> 

Comment the macro is fine.

> Cheers,
> Ahmad
> 
> --
> Pengutronix e.K.                           |                             |
> Steuerwalder Str. 21                       |
> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.
> pengutronix.de%2F&amp;data=04%7C01%7Cpankaj.gupta%40nxp.com%7C4
> d60f0d524a04b7cbd7b08da0d7e7d21%7C686ea1d3bc2b4c6fa92cd99c5c30163
> 5%7C0%7C0%7C637837134158793951%7CUnknown%7CTWFpbGZsb3d8eyJWI
> joiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3
> 000&amp;sdata=PetvZm8teusBwQ4BeZ1VLEOvBlCrZ2k2bNG3SJBEXPw%3D&
> amp;reserved=0  |
> 31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
> Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |