C++ version - my first release ever!

[ifknot]

https://github.com/ifknot/AES-CPP17/releases/tag/v0.1.1-beta

[kokke]

Hi @ifknot and congratulations on the release :)

I haven't had the time (yet) to take a good look at your implementation, but it seems it would be a good match for desktop applications wanting AES encryption.

Have you made any performance-optimizations compared to this version? There are a lot of low-hanging fruits to pick here (issue + PR section has many), that I haven't picked, because it could cost a byte or two in code size.

By the way, thanks for participating in the issues section - I saw you responding to #141 :)

[ifknot]

Hi @kokke thanks my implementation is for fun and learning - the only optimization is loop unrolling as I wondered if manual loop unrolling offered any benefit over compiler optimization e.g. CLang -Ofast and surprisingly it does seem to: (on my system ymmv) https://github.com/ifknot/AES-CPP17/wiki And as you suggested a direct performance comparison kokke vs ifknot shows in favour ifknot with CLang: (on my system ymmv) https://github.com/ifknot/AES-CPP17/wiki/Perfomance-Comparison-Kokke The perfomance tests and R script are in the "tests" folder. I am waiting for a Pi Zero to try cross compiling to in order to compare performance there and memory footprints - I'm hoping that using the Template Factory pattern in C++ should mean a small memory footprint? Also I cant find the NIST advice on xcrypt for CTR mode that you mentioned? Thanks for your interest 🙂

…

________________________________ From: kokke <notifications@github.com> Sent: 14 June 2019 11:13 To: kokke/tiny-AES-c Cc: ifknot; Mention Subject: Re: [kokke/tiny-AES-c] C++ version - my first release ever! (#139) Hi @ifknot<https://github.com/ifknot> and congratulations on the release :) I haven't had the time (yet) to take a good look at your implementation, but it seems it would be a good match for desktop applications wanting AES encryption. Have you made any performance-optimizations compared to this version? There are a lot of low-hanging fruits to pick here (issue + PR section has many), that I haven't picked, because it could cost a byte or two in code size. By the way, thanks for participating in the issues section - I saw you responding to #141<#141> :) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#139?email_source=notifications&email_token=ABBU2HB7N6NH3JYJLREC4B3P2N4MDA5CNFSM4HWJOTXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXWPR2A#issuecomment-502069480>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ABBU2HA5BLHZY3GWKZ4LA63P2N4MDANCNFSM4HWJOTXA>.