06:02:53 <moneronoob345345> hi, whats the chanel for monero mining?
06:45:34 <br-m> <neromonero1024> #xmrmine:matrix.org
16:31:44 <br-m> <sgp_> "Monero FCMP++ Cryptography Implementation Audited by Trail of Bits": https://magicgrants.org/2026/08/17/Monero-FCMP-Cryptography-Implementation-ToB
16:40:36 <br-m> <jpk68:matrix.org> "While we believe that the current implementation is sound on the platforms it targets, it is
16:40:36 <br-m> <jpk68:matrix.org> extremely fragile, and the performance benefits are unclear compared to calling
16:40:36 <br-m> <jpk68:matrix.org> conversion functions to and from a [u8;32] array."
16:40:55 <br-m> <jpk68:matrix.org> Is there any particular reason a pointer is still used here rather than an array?
17:01:31 <br-m> <sgp_> That’s a question for Berman :)
17:37:52 <br-m> <jberman> We'd have to use a distinct internal type for all Points too (which reuse the Scalar type internally), and then call the relatively inconsequential suggested "lightweight" conversion function to convert types that doesn't seem to achieve any significant safety guarantee from my perspective. If the C side writes to the bytes, t [... too long, see https://mrelay.p2pool.observer/e/5NWx0qELd0VXalFG ]
17:39:46 <br-m> <jpk68:matrix.org> Makes sense, thanks
19:08:47 <br-m> <basses:matrix.org> @sgp_: blog post on getmonero?
19:32:16 <br-m> <sgp_> I can write one up if that makes sense
19:46:00 <br-m> <basses:matrix.org> @sgp_: if there are other audits to be done, then I think one big blog about these audits is better
20:19:54 <br-m> <sgp_> I personally think one post about all the audits would be too difficult to follow