[gnutls-devel] GnuTLS | Add Linux kernel AF_ALG backend (!1404)

[Read-only notification of GnuTLS library development activities]

Stephan Mueller commented on a discussion on lib/accelerated/afalg.c: https://gitlab.com/gnutls/gnutls/-/merge_requests/1404#note_552438883

>  		goto end;
>  	}
>  
> -	/* The kernel may set the AAD, avoid modification of auth */
> -	iov[0].iov_base = authtmp;
> -
>  	/* Set CT buffer to be filled by kernel */
> -	iov[1].iov_base = encr;
> -	iov[1].iov_len = plain_size + tag_size;
> +	iov[0].iov_base = encr;
> +	iov[0].iov_len = (encr_size > outbuflen) ? outbuflen : encr_size;
>  
> -	if (kcapi_aead_stream_op(ctx->handle, iov, 2) < 0) {
> +	if (kcapi_aead_stream_op(ctx->handle, iov, 1) < 0) {

Apologies if I am not using gitlab right :-)

But, I see that there is one IOVEC with the encryption buffer used for the _op operation. If I see that right, the encr buffer should be as large to hold the ciphertext || tag data. What I am trying to say is that the kernel, however returns the AAD first, followed by the ciphertext || tag. So, If I would understand it right, the encr buffer is filled with AAD concatenated with parts of the CT and potentially the tag depending on the size of the AAD.

-- 
Reply to this email directly or view it on GitLab: https://gitlab.com/gnutls/gnutls/-/merge_requests/1404#note_552438883
You're receiving this email because of your account on gitlab.com.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.gnupg.org/pipermail/gnutls-devel/attachments/20210415/89aeb1af/attachment.html>