> > > + memset(ipad + keylen, 0, blocksize - keylen);
> > > + memcpy(opad, ipad, blocksize);
> > > +
> > > + for (i = 0; i < blocksize; i++) {
> > > + ipad[i] ^= 0x36;
> > > + opad[i] ^= 0x5c;
> >
> > What are these constant ?
>
> They are defined in the HMAC RFC, as ipad and opad values. See
> https://www.ietf.org/rfc/rfc2104.txt.
>
Since many driver use them, I think defining them in include/ should be done (HMAC_IPAD/HMAC_OPAD)
I will send a patch for it.
> > [...]
> > > +struct safexcel_alg_template safexcel_alg_sha256 = {
> > > + .type = SAFEXCEL_ALG_TYPE_AHASH,
> > > + .alg.ahash = {
> > > + .init = safexcel_sha256_init,
> > > + .update = safexcel_ahash_update,
> > > + .final = safexcel_ahash_final,
> > > + .finup = safexcel_ahash_finup,
> > > + .digest = safexcel_sha256_digest,
> > > + .export = safexcel_ahash_export,
> > > + .import = safexcel_ahash_import,
> > > + .halg = {
> > > + .digestsize = SHA256_DIGEST_SIZE,
> > > + .statesize = sizeof(struct safexcel_ahash_export_state),
> > > + .base = {
> > > + .cra_name = "sha256",
> > > + .cra_driver_name = "safexcel-sha256",
> > > + .cra_priority = 300,
> > > + .cra_flags = CRYPTO_ALG_ASYNC |
> > > + CRYPTO_ALG_KERN_DRIVER_ONLY,
> >
> > Why do use CRYPTO_ALG_KERN_DRIVER_ONLY ?
>
> See http://lxr.free-electrons.com/source/include/linux/crypto.h#L97.
>
Sorry, I had understood that flag as "do not let userspace use me".
Anyway, this flag is totally ignored by the cryptoAPI.
