Buffer changes about af_ktls HOT 6 CLOSED

ad 1)
Yes, this should be dropped for sure. It also limits using kernel_sendpage() (#22 and 0fbc8d1)

ad 2)
Applicable only for data records I suppose. However this will limit switching from OpenSSL, GnuTLS or another TLS lib. Once AF_KTLS socket will be instantiated, switching back to the lib will be possible only if there is some error during decryption or non-data record. If we do not introduce a mechanism for explicitly switching this off. Do we want it?

ad 3)
Sounds good to me.

from af_ktls.

2. Right, only pull them off for data records. Yes, we'd have to figure out something reasonable to do to switch back and forth between ktls and OpenSSL / GnuTLS handling of data. This is similar to #53

from af_ktls.

2. Right, only pull them off for data records. Yes, we'd have to figure out something reasonable to do to switch back and forth between ktls and OpenSSL / GnuTLS handling of data. This is similar to #53

Correct. In the case of openconnect VPN, we would like to switch to userspace handling when a particular application-data packet is encountered (the particular determined by a future feature, to add bfp). If we could do the packet processing in peek mode, and then discard if successful, we would address such a future use case.

from af_ktls.

There were some patches posted upstream that should help with some of the stream parsing

http://comments.gmane.org/gmane.linux.network/420289

from af_ktls.

Mostly fixed by #62, at least the first and second

from af_ktls.

3. fixed by #84

from af_ktls.