niedz., 22 lip 2018 o 15:52 npzacs <g...@videolan.org> napisał(a): > > libaacs | branch: master | npzacs <npz...@gmail.com> | Sun Jul 22 16:33:29 > 2018 +0300| [e2addd80757a28aa75dec9bdf1dc7f17a5a5db3f] | committer: npzacs > > mmc: issue a warning when using AACS2 capable drive
Recently I've bought a LG WH16NS60 drive, which is official AACS2 drive. I've tested a couple of UHD discs and it seems that those drives do not work in a way this patch checks for AACS2. First of all, the 0x010d drive feature seem unchanged. It still reports AACS version 1, no matter what disc is inserted. With AACS1 disc the feature has "current" bit set, with AACS2 disc "current" is unset. I was not able to find a AACS2 drive feature. There is a new feature 0x010f, which is the same size as 0x010d. When I've checked its contents it seem to have 2 in the same place aacs_version is located in 0x010d. It also has 2 in the number of AGID fields, which also seems correct. However, it has "current" bit always cleared, no matter what disc I insert. So I am unsure if this is a feature related to AACS2. Second, I did get some success :) After changing key class field from 2 to 3 it seems I can get a valid AGID and retrieve AACS2 drive certificate. Drive certificate is now 144 bytes long and validates against AACS2 public keys. It seems to have following structure: - 0x11 - probably just cert type, similar to AACS1 - 0x01 - BEC bit - 0x0090 - length - 6 byte drive ID - 6 bytes of zeroes - probably "reserved" - 64 bytes - drive public key - 64 bytes - signature So it seems the above patch won't detect AACS2 drives. At least not this model ;) -- Janusz Dziemidowicz _______________________________________________ libaacs-devel mailing list libaacs-devel@videolan.org https://mailman.videolan.org/listinfo/libaacs-devel
