Re: [dm-crypt] cryptsetup problem on Raspberry Pi 2 w 512bit key-size (works on Raspberry Pi 1, x86_64, 256bit)

[Milan Broz <gmazyland@gmail.com>]

On 02/24/2015 12:02 AM, Johannes Ernst wrote:
> All in one place:
> 
>> dd if=/dev/zero of=./test.img count=8 bs=1M
> 
>> cryptsetup --hash sha512 --key-size 512 -v luksFormat ./test.img
> Used password ‘asdf’ (no quotes)
> 
>> cryptsetup open test.img test
> Enter passphrase for test.img: 
> No key available with this passphrase.
> Enter passphrase for test.img: 
> No key available with this passphrase.

I was able to reproduce it on Raspberry Pi2 B+ and indeed it is Neon code issue.
I am using Raspian, where all cryptocode is compiled as modules:

Without ARM aes_arm_bs module it works:

# echo xxx|src/cryptsetup open test.img --test-passphrase -v
Key slot 0 unlocked.
Command successful.

# modprobe aes_arm_bs

# echo xxx|src/cryptsetup open test.img --test-passphrase -v
device-mapper: reload ioctl on  failed: No such file or directory
Failed to setup dm-crypt key mapping for device test.img.
Check that kernel supports aes-xts-plain64 cipher (check syslog for more info).
Command failed with code 5: Input/output error

(Userspace crypto wrapper failed so code tries to use old dmcrypt temporary device,
errors above are not directly related to Neon issue.)

And in dmesg module even failed test:

[  318.445489] alg: skcipher: Chunk test 1 failed on decryption at page 0 for xts-aes-neonbs
[  318.445515] 00000000: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f
[  318.445526] 00000010: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f
[  318.445534] 00000020: 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f
[  318.445543] 00000030: 30 31 32 33 34 35 36 37 38 39 3a 3b 3c 3d 3e 3f
[  318.445553] 00000040: 40 41 42 43 44 45 46 47 48 49 4a 4b 4c 4d 4e 4f
[  318.445562] 00000050: 50 51 52 53 54 55 56 57 58 59 5a 5b 5c 5d 5e 5f
[  318.445570] 00000060: 60 61 62 63 64 65 66 67 68 69 6a 6b 6c 6d 6e 6f
[  318.445580] 00000070: 70 71 72 73 74 75 76 77 78 79 7a 7b 7c 7d 7e 7f
[  318.445588] 00000080: 80 81 82 83 84 85 86 87 88 89 8a 8b 8c 8d 8e 8f
[  318.445597] 00000090: 90 91 92 93 94 95 96 97 98 99 9a 9b 9c 9d 9e 9f
[  318.445607] 000000a0: a0 a1 a2 a3 a4 a5 a6 a7 a8 a9 aa ab ac ad ae af
[  318.445616] 000000b0: b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 ba bb bc bd be bf
[  318.445624] 000000c0: c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 ca cb cc cd ce cf
[  318.445634] 000000d0: d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 da db dc dd de df
[  318.445643] 000000e0: e0 e1 e2 e3 e4 e5 e6 e7 e8 e9 ea eb ec ed ee ef
[  318.445651] 000000f0: f0 f1 f2 f3 f4 f5 f6 f7 f8 f9 fa fb fc fd fe ff
[  318.445661] 00000100: 00 01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f
[  318.445670] 00000110: 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f
[  318.445678] 00000120: 20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f
[  318.445688] 00000130: 30 31 32 33 34 35 36 37 38 39 3a 3b 3c 3d 3e 3f
[  318.445697] 00000140: 40 41 42 43 44 45 46 47 48 49 4a 4b 4c 4d 4e 4f
[  318.445705] 00000150: 50 51 52 53 54 55 56 57 58 59 5a 5b 5c 5d 5e 5f
[  318.445715] 00000160: 60 61 62 63 64 65 66 67 68 69 6a 6b 6c 6d 6e 6f
[  318.445723] 00000170: 70 71 72 73 74 75 76 77 78 79 7a 7b 7c 7d 7e 7f
[  318.445732] 00000180: 80 81 82 83 84 85 86 87 88 89 8a 8b 8c 8d 8e 8f
[  318.445742] 00000190: 90 91 92 93 94 95 96 97 98 99 9a 9b 9c 9d 9e 9f
[  318.445750] 000001a0: a0 a1 a2 a3 a4 a5 a6 a7 a8 a9 aa ab ac ad ae af
[  318.445759] 000001b0: b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 ba bb bc bd be bf
[  318.445769] 000001c0: c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 ca cb cc cd ce cf
[  318.445777] 000001d0: ab 1c bb 4c 15 50 be 97 f7 ab 40 66 19 3c 4c aa
[  318.445786] 000001e0: 5b 41 ef c3 ba 09 50 93 f2 c5 00 48

Seems there is an issue with the Neon AES implementation...

Milan