@UkoeHB the methods wallet2::get_multisig_seed() and wallet2::restore() (the multisig overload), haven't been touched in 5 years and do not store/load the field m_multisig_derivations. I might be wrong, but it's probably necessary to use this field to restore multisig wallets from scratch (and skip the initial key exchange / preparation stages), yeah?

* @UkoeHB the methods wallet2::get_multisig_seed() and wallet2::generate() (the multisig overload), haven't been touched in 5 years and do not store/load the field m_multisig_derivations. I might be wrong, but it's probably necessary to use this field to restore multisig wallets from scratch (and skip the initial key exchange / preparation stages), yeah?

* @UkoeHB the methods wallet2::get_multisig_seed() and wallet2::generate() (the multisig overload), haven't been touched in 5 years and do not store/load the field m_multisig_derivations. I might be wrong, but it's probably necessary to use this field to restore multisig wallets from scratch, skipping the initial multisig setup, yeah?

Nvm it seems the derivations aren't needed after finalization

jeffro256[m]: right

How many multisig private keys should there be given M/N?

(after setup)

Koe could you pretty please double check that monero-project/monero #8914 is correct?

jeffro256[m]: one for every group of size N - M + 1

so I guess N choose (N - M + 1) would do it (there is n choose k in seraphis_lib src/seraphis_crypto/math_utils.h)