02:48:22 Does the arm64 build compile supercop? If so, how? 02:48:24 If not, then how? (I’m rly confused, I can’t figure out how CMake skips the library for non amd64 platforms) 03:26:53 Yeah uh I think I figured it out 03:26:54 Doesn’t mean I like it though 03:27:50 If I’m not mistaken, it isn’t currently possible to compile anything dependent on supercop on MacOS x86 targeting MacOS Arm 03:28:34 At least not without having to create an entire cross compilation toolchain….*which is wholly unnecessary on MacOS for this use case* 03:29:37 I’ll look at this l8r tmr, it isn’t *too* hard to fix, just need to change the CMAKE_SYSTEM_PROCESSOR check 10:28:44 preland: are you working on this? https://github.com/monero-project/monero/pull/9357 (universal macOS binaries) 11:58:34 Sort of 11:58:34 I’ve been ramming my head against a wall trying to build monero-java as a universal Mac for a while now, and that was from one of my earlier attempts 11:59:43 I think I’ve finally figured out how I want to do it, but there are still random roadblocks everywhere 11:59:48 Yes 14:53:25 it's not clear to me how you are trying to solve this, you would need to update depends/gitian build scripts 14:53:45 but i find the benefits of universal binaries questionable, it would mean larger download sizes 15:17:39 preland @preland:monero.social: supercop only has x86-64 acceleration right now. There's a neon implementation, but I think it isn't compatible with arm64 abi. There's a 64bit floodberry implementation - which isn't a part of supercop - that's probably faster but hasn't been merged for use by the wallet 15:47:45 Yeah supercop doesn’t work for non x86_64 atm 15:47:46 When I meant “fix” I meant that when I compile for Arm on Mac x86 it would see it as Arm; it currently sees it as x86, which causes it to attempt to compile supercop as arm…. 15:50:40 It’s not surprising that supercop doesn’t check for that Apple edge case, as it is older than MacOS arm