65 replies to this topic

[SuicidalSTDz]

Engineer, on 23 December 2014 - 10:24 PM, said:

Can I please be taken off the list as developer? Thats because I do not code for ComputerCraft anymore, and I want this topic to be at the very least accurate
Also, it is nice to see this alive again!

Of course! Nice to be back in business

[Anavrins]

About the crypto api, specifically the AES part, I tried testing it against the official FIPS test vectors and they did not pass.
After some research, it looks like the key_schedule part is not properly implemented.
https://dl.dropboxus...37842828/cc.png

[SuicidalSTDz]

Anavrins, on 31 December 2014 - 07:12 AM, said:

About the crypto api, specifically the AES part, I tried testing it against the official FIPS test vectors and they did not pass.
After some research, it looks like the key_schedule part is not properly implemented.
https://dl.dropboxus...37842828/cc.png

I'll speak to KillaVanilla about fixing that. It's not exactly my cup of tea.

[KillaVanilla]

SuicidalSTDz, on 31 December 2014 - 07:12 PM, said:

Anavrins, on 31 December 2014 - 07:12 AM, said:

About the crypto api, specifically the AES part, I tried testing it against the official FIPS test vectors and they did not pass.
After some research, it looks like the key_schedule part is not properly implemented.
https://dl.dropboxus...37842828/cc.png

I'll speak to KillaVanilla about fixing that. It's not exactly my cup of tea.

Anavrins, on 31 December 2014 - 07:12 AM, said:

About the crypto api, specifically the AES part, I tried testing it against the official FIPS test vectors and they did not pass.
After some research, it looks like the key_schedule part is not properly implemented.
https://dl.dropboxus...37842828/cc.png

I'm looking into it, don't worry.

[Anavrins]

In addition to the Key Schedule, the shiftRows, invShiftRows and mixColumns functions were also wrong.
I made a pull request to fix them
https://github.com/S...nderAPI/pull/54

Edited by Anavrins, 16 January 2015 - 07:33 AM.

[Anavrins]

I figured out the problem with the Key Schedule, there was also an error with the round operations.
I've added both of these fix to my pull request, hope you consider it.
After testing AES again with all these fixes, it now correctly pass the test vectors!

Edited by Anavrins, 20 January 2015 - 05:22 AM.