16:51:25 Meeting in a bit more than 1 hour 18:00:04 Meeting time. Hello! https://github.com/monero-project/meta/issues/1258 18:00:17 hello 18:01:15 howdy 18:01:46 hi 18:01:53 *waves* 18:02:15 Alright, onwards to the reports from last week 18:02:51 made great improvements on the Wallet APIs `wallet_keys_unlocker` thanks to jeffro 18:02:53 also finally fixed the bug that made `verifyPassword()` fail after make_multisig 18:03:36 So the hunt for the last remaining pieces of direct wallet2 use in the CLI wallet can continue :) 18:04:11 yay, way more fun than starring at the debugger :D 18:04:38 If the bug is interesting ... 18:06:08 me: fixed a couple bugs testing forking from current testnet (in the migration and in scanning) in prep for alpha stressnet, cleaned up the FFI (removed unwraps/asserts, used int error returns to gracefully handle errors, de-duplicated some code with macros, clippy+fmt), implemented consolidated paths in the RPC for getting paths in the tree by global output id, started on a bette r organizational structure for curve trees logic in the db code, tested kayaba's latest prove/verify optimizations (good results!!) 18:06:11 Planning to complete organizational structure in the next day, then moving to documentation/Carrot review, in prep for opening PR's to the main Monero repo 18:06:35 Good news: kayabanerve implemented linear prove() times, dropping 128-input tx construction down from 5m30s to ~1m!! Huge 18:06:44 me: initiated communication about Carrot follow-up review by Cypherstack, updated FCMP++ benchmark tool for all the latest FCMP++ changes (github.com/jeffro256/clsag_vs_fcmppp_bench/), working on supporting high input counts in benchmark tool, working on patches in Monero core repo, re-reviewing the de-dup PR by rbrunner7, reviewed several PRs in seraphis-migration, did a write-up for an upcoming vuln disclosure, helped prepare v0.18.4.2 release 18:07:34 Rucknium answered jeffro256 's questions that came up during his review, and I made the requested (smaller) changes. I think it's ready now for a second review, that certainly seems a good idea because of the importance of the code: https://github.com/monero-project/monero/pull/9939 18:08:17 "an upcoming vuln disclosure". Oh, always interesting. Did the fuzzing already turn up something auctionable? 18:09:00 👋 18:09:31 I deduplicated 25k lines of code and optimized FCMP++ prove. Apologies for being a few minutes late to the meeting. 18:10:33 That's the 5 times speedup that jberman mentioned? 18:10:36 Yes, but this specific one isn't related to the fuzzing group. The report will go up as soon as v0.18.4.2 binaries are out, not sooner so users don't get confused. 18:10:39 I also raised the topic of migrating the hash to point algorithm with FCMP++s after finally successfully identifying the present algorithm. 18:10:43 Largely rbrunner7 18:11:24 But if you're self-compiling for your own wallet, just make sure to be on the latest release-v0.18 or master branch. 18:11:43 Just curious, what is there to win if we migrate to another "hash to point" algorithm? A bit faster? 18:11:57 jeffro256: Thanks for the hint 18:12:02 Bits of security currently being lost. 18:12:10 jeffro256 would this be on -site as a blog post also - pushed at the same time binaries are? 18:12:29 We currently use keccak256. Would you be happy if I set the first byte to 0 after every usage? 18:12:49 We currently use a hash to point which has an explicit bit of bias and probably has further bias after. 18:12:55 Is this a trick question :) 18:13:00 a new hash to point just tightens everything back up 18:13:33 It's also only ~10 lines of cryptography. It just invokes the existing hash to curve twice. It isn't a major change re: cryptography 18:13:57 And no hairy compatibility issues if we switch that algorithm? Or will it be done in standard version based way? 18:14:21 Before like so, afterwards differently 18:14:49 plowsof: yes that could be done 18:15:05 New outputs are added to the tree with a key image generator sampled using the new hash to point 18:15:07 Old outputs are added to the tree with their key image generators sampled using the existing hash to point 18:15:09 The world keeps spinning 18:15:20 Alright :) 18:16:49 It's a type of vuln that doesn't really benefit much from being hidden from attackers since it's a data leak, it mainly just hurts users the longer they don't update. If it makes sense, perhaps a report PR can be made to the monero-site, but I just don't want it to spread on social media sites before the release binaries are posted, and cause confusion 18:17:14 @selsta 18:17:24 plowsof: bumping this PR too not sure who else to bump to: https://github.com/monero-project/monero-site/pull/2510 18:19:33 Ok. Anything else to discuss today beyond these reports? 18:21:00 Doesn't look like it. I think we can close. Thanks everybody for attending, read you again next week! 18:21:31 fuzzing found issues but so far none have been exploitable 18:21:56 thanks everyone 18:22:20 Good to hear, selsta. 18:24:25 By the way, SNeedlewoods , it looks as if eigenwallet really went the API way instead of using wallet2 directly, which is nice: https://libera.monerologs.net/monero/20250712#c540320 18:24:55 Yes, sorry, I should have been more clear about the changes made due to the fuzzing group: actionable, but not exploitable. 18:25:14 Thanks everybody 18:25:15 (There must be some more recent statement, but can't find right now) 18:26:51 had a chat with the devs a while ago and gave them a link to the branch I'm currently working on, so they can track the most recent changes to the API 18:27:18 rbrunner7: about the changes? 18:27:39 for fuzzing? 18:27:41 Er, no, about eigenwallet using the API 18:28:10 Oh okay nvm