17:00:38 <br-m> <rucknium> Meeting time! https://github.com/monero-project/meta/issues/1488
17:00:40 <br-m> <rucknium> 1. Greetings
17:00:44 <br-m> <vtnerd> hi
17:00:45 <br-m> <jberman> waves
17:00:59 <br-m> <syntheticbird> hello
17:01:00 <br-m> <jpk68:matrix.org> Hello
17:01:01 <br-m> <sgp_> hello
17:01:10 <tevador> Hi
17:02:30 <br-m> <jeffro256> Howdy
17:03:06 <br-m> <rucknium> 2. Updates. What is everyone working on?
17:03:19 <br-m> <jpk68:matrix.org> Me: working on replacing the HTTP server in epee, cross-platform hardening for the GUI
17:04:00 <br-m> <rucknium> me: Wrote "Research directions for selfish mining countermeasures in Monero" https://gist.github.com/Rucknium/0b6e94811099bfab44e84562621d8fe1
17:04:10 <br-m> <jeffro256> Testing for InProofs/OutProofs for carrot_core, core_tests and updates for #10916  
17:04:11 <br-m> <vtnerd> me: updated lws fcmp++ code, in the process of a merge/update to lwsf fcmp++ code, and working on (hopefully) last round of ssl edits. the ssl patch is looking pretty good now (another push later today likely)
17:05:49 <br-m> <jberman> me: followups to FCMP++ integration PR's (FCMP++ verify tooling was merged also)
17:12:23 <br-m> <jberman> Integrated tevador 's relative lock in the FCMP++ consensus integration PR also in order to enable trustless payment channels as proposed
17:12:23 <br-m> <rucknium> 3. FCMP++ Phase 2 & 3 integration audits. Table of firm quotes (https://mrelay.p2pool.observer/m/monero.social/cQmCpBmZbUUCgfJTYajCksPm.png).
17:12:24 <br-m> <sgp_> I am very appreciative to the 8 vendors who submitted bids for this integration audit. We received quotes from many competent vendors, and it has been challenging to pick the final option
17:12:24 <br-m> <sgp_> After much discussion with Berman, I would like to receive approval at this meeting to go forward with Vendor 6. Vendor 6 is highly skilled. I will conditionally request that this Vendor include one highly skilled reviewer who Berman especially wants on the project, and I will ensure that the auditor in their report states tha [... too long, see https://mrelay.p2pool.observer/e/6P_cqL0LNTRrR0t2 ]
17:12:24 <br-m> <sgp_> With those, I think they are the clear choice for the best single reviewer
17:12:24 <br-m> <syntheticbird> is it zellic?
17:12:24 <br-m> <syntheticbird> if its zellic i'm for, if it's not zellic im against
17:12:24 <br-m> <jpk68:matrix.org> Is it Aumasson?
17:12:25 <br-m> <sgp_> @syntheticbird: They did not submit a bid for this one
17:12:49 <br-m> <syntheticbird> @sgp_: fml
17:14:06 <br-m> <syntheticbird> more seriously, if berman is backing one of them there, no reason to refuse
17:14:36 <br-m> <jberman> Adding, the reason I liked that specific reviewer is that reviewer in particular identified a very relevant integration inflation vulnerability in another project that is of a class similar to what would be reviewed in this audit. I also found no known integration inflation misses on any of their past audits too
17:16:27 <br-m> <sgp_> at today's prices, the CCS has about $218k remaining. So in theory another option (Vendor 1?) could also be chosen, but there probably is good reason to save the money as well
17:17:09 <br-m> <syntheticbird> what will be done of the leftovers after FCMP++ and Carrot goes mainnet
17:17:36 <br-m> <syntheticbird> actually that's a question maybe worth answering after consensus is reached on vendor 6
17:18:09 <br-m> <jberman> We had ideas on followup audits: mx25519 audit, prover side audits, academically proving the faster torsion check. The funds could be allocated toward any of those
17:18:31 <br-m> <sgp_> fwiw I tried to justify picking multiple vendors in combination for around the same price as Vendor 6, but I can't justify any combination instead of Vendor 6. I think their quote is clearly the best
17:18:49 <br-m> <rucknium> @jberman: By my count, the auditors/reviewers that missed a counterfeiting flaw in a blockchain that had the flaw discovered or exploited this year (2026) are: HashCloak, Linfeng(Daniel) Zhou, Least Authority, Kudelski, SmartDec, qedit, nccgroup, Quarkslab, Inversed Tech
17:19:24 <br-m> <jeffro256> I considered combining quotes, but I think that at least for a first passthrough, we should prefer depth to quantity 
17:20:18 <br-m> <sgp_> @rucknium: I would love a table of this if you can put it together sometime, to make sure I didn't "miss" one
17:20:35 <br-m> <rucknium> These involved Firo, Beam, Zcash, LTC MWEB, and Iron Fish. Some of the other vulnerable blockchains this year did not have a formal independent audit.
17:21:20 <br-m> <rucknium> @sgp_:monero.social: I can probably put one up in the next day or two.
17:22:36 <br-m> <sgp_> Just to get it on all y'alls radar, the next RFQ (relating to kayaba's research CCS) should relate to multisig formalization and review. Obviously no action needed today for that
17:23:17 <rbrunner> I guess that very low offer from vendor is not considered really credible? Looks like a misunderstanding somewhere, compared with the others
17:23:24 <rbrunner> *vendor 3
17:24:04 <br-m> <rucknium> I support Vendor 6.
17:24:52 <br-m> <rucknium> We want to acheive loose consensus here today. Please ask questions and raise any objections ASAP.
17:25:05 <br-m> <sgp_> We don't recommend the vendor 3's quote. I believe it was the weakest of the ones we received, though I am grateful to have received it all the same
17:25:06 <br-m> <rucknium> achieve*
17:25:50 <br-m> <sgp_> The cheapest review I would recommend is Vendor 1's, but again, in addition to Vendor 6 not instead of
17:26:37 <br-m> <sgp_> the amount of review time for Vendor 4 would be valuable but budget does not allow for 6+4
17:29:20 <br-m> <rucknium> More discussion of this proposal?
17:29:22 <br-m> <gingeropolous> i guess this is loose consensus amongst those that know the details?
17:30:12 <br-m> <jeffro256> Agreed 
17:31:14 <br-m> <gingeropolous> i mean, we are looking for a loose consensus for those that know the details. Not that if i knew the details i could really do much anyway
17:32:04 <br-m> <syntheticbird> that's why i pull the "zellic shiller" card immediately, because I knew I would have nothing else to tell about other vendors even with the details.
17:32:58 <br-m> <rucknium> I see loose consensus in favor of Vendor 6 for the FCMP++ Phase 2 & 3 integration audits.
17:33:34 <br-m> <sgp_> plowsof please help me get luigi on the line :)
17:33:44 <br-m> <sgp_> Ty
17:35:05 <br-m> <jberman> Of note, we'll want 10916 in and finalized before giving them final commit hashes to audit for Phase 3. Phase 2 should be good to go
17:35:52 <br-m> <rucknium> Congratulations. Let's move to the next item on the agenda.
17:35:59 <br-m> <rucknium> 4. get_tx_key/check_tx_key doxes sender in 2-out tx (https://github.com/monero-project/monero/issues/11586).
17:36:10 <br-m> <rucknium> Does this need to be discussed more?
17:36:47 <br-m> <jeffro256> No new updates AFAIK 
17:37:28 <br-m> <rucknium> Does it need to stay on the agenda for next week?
17:37:31 <br-m> <jeffro256> It can be removed from the agenda if no one else has anything to say about it 
17:37:58 <br-m> <rucknium> 5. FCMP++ to-do list status. Programming tasks (https://github.com/seraphis-migration/monero/issues/53). Reviews and audits (https://cryptpad.fr/sheet/#/2/sheet/view/yPVIUywwA9-deE9VF6GYm9bXbPdCerdST3UDEEfBxcM/embed/). FCMP++ Integration Audit Overview (https://github.com/seraphis-migration/monero/issues/294). Network upgrade  [... too long, see https://mrelay.p2pool.observer/e/4s-9qb0LbkdhTEx5 ]
17:39:12 <br-m> <jberman> FCMP++ integration PR's coming along (verify tooling and db integration now both merged)
17:39:35 <br-m> <jberman> I'm about to move on to reviewing hinto's PoWER impl (finally)
17:41:16 <rbrunner> Good news :)
17:41:34 <br-m> <rucknium> More on this topic?
17:42:32 <br-m> <rucknium> 6. FCMP beta stressnet (https://github.com/seraphis-migration/monero/releases/).
17:43:15 <br-m> <jberman> No developments to report on stressnet since last week on my end
17:43:37 <br-m> <jeffro256> Same 
17:44:10 <br-m> <rucknium> 7. Share or Perish (selfish mining mitigation) (https://github.com/monero-project/research-lab/issues/146).
17:44:35 <br-m> <rucknium> I wrote up my thoughts on where to take selfish mining research: https://gist.github.com/Rucknium/0b6e94811099bfab44e84562621d8fe1
17:44:45 <br-m> <rucknium> Feel free to disagree!
17:45:54 <br-m> <rucknium> In summary, use Markov Decision Process (MDP) to compare available options, including Share or Perish. Set the adversary's objectives to be the traditional revenue maximization and the nontraditional disruption maximization.
17:46:28 <br-m> <rucknium> @gingeropolous:monero.social: ^ You can stick that in the bot's pipe and smoke it.
17:46:55 <br-m> <gingeropolous> indeed
17:47:26 <rbrunner> Those events were interesting enough to get other people to write papers about? Wow.
17:47:45 <br-m> <rucknium> rbrunner: What do you mean? Qubic?
17:48:32 <br-m> <rucknium> Qubic is really the only documented episode of selfish mining. There are so many theoretical papers about selfish mining, but until now so little evidence that it actually mattered in the real world.
17:48:47 <rbrunner> Yes
17:48:58 <rbrunner> I see.
17:49:05 <br-m> <rucknium> And Qubic's behavior was not even that of a classic selfish miner. I discuss that in the note.
17:49:52 <br-m> <rucknium> Its goal was more to produce disruption that would convert into publicity for its own blockchain, IMHO.
17:50:45 <br-m> <gingeropolous> from my exploration in monerosim, SoP moves the bar from the .33 to somewhere around .46 , and the connectivity of honest miners matters a lot to how effective it is. 
17:51:21 <rbrunner> Very interesting reading, this note.
17:51:32 <br-m> <rucknium> Makes sense. I discuss the importance of the tiebreaking gamma parameter, that models network propagation.
17:52:29 <br-m> <rucknium> Thanks, rbrunner . I tried to explain why I favor certain methodological choices instead of just saying "Use these methods because I said so."
17:53:28 <br-m> <rucknium> More on this topic?
17:54:34 <br-m> <rucknium> 8.  Shi et al. (2026) "Are Unreachable Nodes Truly Safe? Fully Eclipsing Monero’s P2P Network!" (https://arxiv.org/pdf/2609.10260). p2p: increase white_list size and change connection selection algorithm (https://github.com/monero-project/monero/pull/11527). p2p: deduplicate outbound peers by netgroup and bucket the peer lists (https://github.com/monero-project/monero/pull/11707).
17:55:38 <br-m> <rucknium> With https://github.com/monero-project/monero/pull/11707 , 0xFFFC0000 and @ofrnxmr:monero.social have a proposed PR to defeat the Nyx eclipse attack.
17:55:49 <br-m> <thomasbuilds:matrix.org> I'm not here to discourage (expensive) audits but just know I did many audits and it's basically entirely clean. > <@jberman> We had ideas on followup audits: mx25519 audit, prover side audits, academically proving the faster torsion check. The funds could be allocated toward any of those
17:55:49 <br-m> <thomasbuilds:matrix.org> btw tevador maybe you'd like to know that mx25519's AArch64 code never turns on the CPU's DIT flag so on cores where timing can depend on the data without it (Apple silicon, where it also leaves the M3's GoFetch prefetcher on and the Cortex-A510's 64-bit mul, which only debug builds of the portable code use) its constant-tim [... too long, see https://mrelay.p2pool.observer/e/1f7-qb0LbnZrdEtl ]
17:56:46 <br-m> <ofrnxmr> @rucknium: Needs a couple things addressed, but is mostly finished
17:57:00 <br-m> <rucknium> selsta put my PR and 0xFFFC0000 & ofrnxmr's on the milestone to complete before the next release version (v0.19) https://github.com/monero-project/monero/milestone/2 . Thanks, selsta
17:57:54 <br-m> <rucknium> #11707 includes my two changes to the connection selection algorithm and omits the change to the white_list size. It changes how the white_list is structured.
17:58:07 <br-m> <rucknium> @ofrnxmr:monero.social: Let me know when you want me to run it in monerosim
17:59:35 <br-m> <rucknium> More on this topic?
18:00:10 <br-m> <gingeropolous> im exploring some more antiproxy things but i don't have a coherent story yet
18:01:37 <br-m> <rucknium> We can end the meeting here. Feel free to continue discussing. Thanks everyone.