> >> - No Adiantum in userspace crypto libraries currently supported by > >> libcryptsetup. The fallback via dm-crypt will again need CAP_SYS_ADMIN. > >> Affects LUKS2. > > > > Is there anything blocking addition of these to openssl, or is it merely > > that no one has requested it historically ? (Perhaps naively) I would > > think that openssl ought to be supporting all of the cipher/mode. > > I really can't say for sure. I'd only guess people did not care as there > was an alternative via AF_ALG, at least in libcryptsetup use case. > > But adding new things in libraries take time and a _lot_ of effort. You > can read about 2+ years journey of Argon2 KDF in openssl[1][2], if you'd > like :)
From what I can see with Adiantum it was added to the kernel in 5.0 for Android Go phones running on ARMv7 (Cortex-A7) processors because they had really slow AES, and it wasn't used for DM there. Maybe the communication of it's demise and ensuring people know to migrate rather than enabling further proliferation is the right way to go, in Fedora people would have had to really gone out of their way to use it in cryptsetup. Also it's not going away in F-45 for this usecase so it's more about warning people to prepare to migrate, not forcing it as part of this change. Again better to notify users so they have warning rather than pulling the rug out in a random kernel upgrade in the future. Peter -- _______________________________________________ devel mailing list -- [email protected] To unsubscribe send an email to [email protected] Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/[email protected] Do not reply to spam, report it: https://forge.fedoraproject.org/infra/tickets/issues/new
