3 ms·
Hey, I found a small typo in the presentation. On page 80, in "good" code sample, you can see: x |= MAC_computed[i] − MAC_computed[i]; It probably should
by M4v3R 12y ago
Hey, I found a small typo in the presentation. On page 80, in "good" code sample, you can see:
x |= MAC_computed[i] − MAC_computed[i];
It probably should be something like:
x |= MAC_computed[i] − MAC_received[i];
- KMag 12y agoOr better yet x |= MAC_computed[i] ^ MAC_received[i]; If they're not careful and the numeric type being subtracted is wider than CPU registers, depending on architecture, the compiler-generated carry code to implement wider-than-register subtraction may introduce timing attacks. Wider-than-register xor is much much less likely to have such issues.