00:00:17 <eureka> or well, my account appears to be broken on IdP, can't log in right now but I will get that fixed and then try builds
00:04:58 <eureka> CMake Error at CMakeLists.txt:29 (cmake_minimum_required): Compatibility with CMake < 3.5 has been removed from CMake.
00:05:03 <eureka> $ cmake --version
00:05:03 <eureka> cmake version 4.1.2
00:05:05 <eureka> :think:
00:13:26 <eureka> seems clang does not like the definitions for the vector types; https://astr.al/u/45c1a436.txt
00:28:58 <eureka> fixed typedefs, now some failure to resolve function impl https://astr.al/u/f8d2d793.txt
00:30:16 <m-relay_> <g​ingeropolous:monero.social> ah yes. the [--v2] means with or without
00:37:40 <eureka> altivec.h defs don't like cstdint types, (signed long long) is all it wants
00:40:48 <eureka> ok, builds, tests pass ok https://astr.al/u/015e5ae4.txt
00:45:37 <eureka> https://astr.al/u/acd614e9.txt
00:48:33 <eureka> https://astr.al/u/acd614e9.txt light mode is ok, mine mode seems to be having a very bad time allocating extremely slowly 
00:48:58 <eureka> i do not have root on this machine and there are no largepages so that doesn't help much
00:49:15 <eureka> but in any event, interpreter seems to be ok on power8 BE 
00:58:11 <eureka> and I just notice I did all this in master, not v2 branch. whoops
01:01:48 <DataHoarder> it's interpreter either way, so not much speed loss
01:37:14 <eureka> ok, altivec intrins under clang have different prototype for __builtin_crypto_v{,n}cipher - gcc  vector unsigned long long, clang (and by extension, xlC) vector unsigned char
01:37:24 <eureka> https://www.redbooks.ibm.com/redbooks/pdfs/sg248171.pdf pp 149
01:43:24 <eureka> this becomes a disgusting mess of ifdef .. why does it need to have a different signature :sigh:
17:02:34 <gingeropolous> do you like scripts? i like scriots: https://github.com/Fountain5405/scriptsforthewin/blob/main/randomx_benchmark_full.sh
17:02:41 <gingeropolous> scriots
17:47:48 <sech1> Nice