-
l-mdoes anyone know how i could get in contact with tevador? i would like to create a JS implementation of their EquiX library and would like to ask for permission to swap the LGPL license to a license more suitable for javascript, like MPL 2.0 (to retain the copyleft).
-
sech1Ask on github?
-
l-mi just created an issue. i doubt he will respond however, it seems like he is quite inactive nowadays
-
l-mthats why i asked
-
m-relay<fede:xmr.mx> are you the legend behind RandomX.JS? EquiX JS would be actually a no-brainer! it would have so much usecase (e.g. antispam! DDoS protection! and more!)
-
m-relay<gingeropolous:monero.social> if tevador doesn't return before the fork, can the proposed changes for randomx still move forward?
-
l-mpeople back in 2020 asked about a WASM version of EquiX, tevador hasn't responded since
-
l-mi am iffy about forking and releasing EquiX.js, since LGPL isn't suited for the javascript world as a license. i would prefer MPL
-
l-mi would have to wait for his approval
-
sech1gingeropolous RandomX v2 is almost done, it's only missing RISC-V code. But I can finish it when the time comes. tevador/RandomX #274 has x64 code, and I left a comment there with a link to ARMv8 code
-
hycBtw, the gcc compile farm has just added some RISC-V nodes with RVV1.0 vector support
-
hyc
-
hycI have an account there, lemme know if you want to test any risc-v code there
-
sech1I can test RISC-V on Github CI: github.com/jirutka/setup-alpine (it uses QEMU)
-
sech1But ideally, for a proper risc-v development, a physical dev board is much more convenient. I need to get one eventually.
-
hycqemu, ugh. wait forever for tests to run
-
m-relay<fede:xmr.mx> hey l-m, AFAIK LGPL doesn't apply if you rewrite the project, even if it's just a reimplementation
-
m-relay<fede:xmr.mx> a complete rewrite of EquiX would bypass the license restriction
-
m-relay<fede:xmr.mx> and i doubt tevador will bring you to court because you reimplemented his hash algorithm
-
m-relay<fede:xmr.mx> an efficient wasm implementation of equix (or a similar GPU/FPGA/ASIC resistant algorithm with instant verification times) would really be so cool for security purposes
-
gingeropolous_word, thanks sech1 . Do you think it'll need any extensive audits? or is it more little things that have been modded
-
sech1It's not a massive change in RandomX v2