Comment 18 for bug 1111882

TJ stated in early 2013:

"Additional research seems to indicate this is a known intentional gnutls behaviour (that has been modified in very recent gnutls that makes use of a recent libnettle - as mentioned above). The issue is, apparently, the random size padding of packets to prevent communications compromise for stream ciphers.

Unfortunately the changes required are far too invasive for an SRU so we'll have to make do with a work-around."

Where do these changes need be implemented? In GNUTLS?
Has anyone with sufficient understanding opened an informed report upstream at https://gitlab.com/gnutls/gnutls/issues that could be referenced here?
If not, would someone be willing / capable to do that?

Thanks!