03:15:56
ofrnxmr:
@tomdooley:matrix.org: They'll be or have been made
04:38:30
DataHoarder:
I got it running already on stressnet for go-p2pool and changed already for next beta
04:39:12
DataHoarder:
There were some questions and requests and some were implemented to benefit (or allow working) of p2pool and others to miners in general
04:40:32
DataHoarder:
C p2pool is not updated but that'd be simpler with changes known (and documented). Schemes that use FCMP++ ability to chain transactions to aggregate outputs automatically are in research (but don't necessarily are required for the upgrade)
04:41:09
DataHoarder:
xmrig was updated for stress net already, but if mining via p2pool you had to do nothing to have it work for now
05:15:25
DataHoarder:
the only change that was made to p2pool (C-p2pool) ahead of time and is live now is checking for torsion on wallet addresses and bailing out client-side, for the upgrade ahead of time they will be refused network-wide
08:58:48
sech1:
It will be updated for FCMP++ when the time is right. Right now it's too early. Updating for FCMP++ is no small feat, in fact it's a massive change. Definitely not something for AI to do, no matter how good it is.
09:28:45
DataHoarder:
oh right it was tomdooley. No that's a super specific field knowledge that AI will never do right
09:29:49
DataHoarder:
the "p2pool" changes are small, relatively. However they depend on bringing all the carrot crypto into p2pool ^ which I did on my project
09:29:56
DataHoarder:
(which is also a moving target)
09:32:13
sech1:
FCMP++ changes must be tied together with RandomX v2 changes which are also part of the next upgrade, and both of those changes will require a block template/header format change. This is not finalized yet, so why I said it's too early.
09:33:15
sech1:
and carrot crypto, yes
10:53:24
DataHoarder:
it's all a moving target until final release after all
10:54:02
DataHoarder:
I make the changes myself for early sanity checks (hey, we got a few tips of things to modify) but I had to change that code several times as the target indeed moves and shifts
15:04:39
rucknium:
MRL meeting in this room in two hours.
16:59:39
rucknium:
Meeting time! https://github.com/monero-project/meta/issues/1340
16:59:42
rucknium:
1. Gtreetings
17:00:05
rbrunner:
Hello
17:01:27
jberman:
waves
17:01:41
articmine:
Hi
17:02:55
rucknium:
2. Updates. What is everyone working on?
17:05:11
jberman:
me: PR followup, set up a 4-phase plan to get the FCMP++ integration audited, got pre-fork multisig working / tests passing
17:06:20
rucknium:
3. FCMP code integration audit prep.
17:07:47
syntheticbird:
Hi
17:08:25
jberman:
My current plan on the table is to have the fcmp++-stage in the seraphis-migration repo ready for auditing. I'd like to audit the integration code in 4 phases: 1) Crypto, 2) Crypto Integration, 3) Tree building / prove / verify, 4) Consensus integration
17:09:32
jberman:
Each subsequent phase essentially utilizes the building blocks of its preceding phase
17:09:50
vtnerd:
Hi
17:10:36
jberman:
So right now, I have PR's prepped (and am working with jeffro to get those PR's merged), and then would like to get started with auditing on a specific commit of fcmp++-stage
17:11:02
vtnerd:
Sorry for late followup: me: testing changes to boost beast in lwsf. Getting monero_c/lwsf to work on macos should be completed
17:11:32
rbrunner:
So with possible working-in of the results of one phase's audit results before going to the audit of the next phase?
17:11:38
jberman:
Of note, I've already opened a couple PR's upstream, which jeffro and vtnerd reviewed. I'm proposing we get that code + other related building block crypto audited as well
17:12:32
jberman:
rbrunner: Yes. I'm thinking about a distinct CCS like kayaba's where I raise funds for the audits in advance, to minimize downtime between each phase
17:13:02
rbrunner:
I see.
17:13:08
rbrunner:
Makes sense
17:14:02
rbrunner:
Will be interesting to see how long the whole process takes, over all 4 phases
17:14:16
jberman:
Here are the specific sections I want to get audited: https://paste.debian.net/hidden/82c00500
17:15:47
jberman:
rbrunner: I think 3 months is a reasonable expectation. First 2 audit phases ~1 month, and the latter 2 ~two months
17:16:21
rbrunner:
Sounds like a good sprint :)
17:16:46
rbrunner:
Hopefully without burnout at the end ...
17:20:08
rucknium:
Do you have specific firm(s) in mind?
17:20:50
jberman:
Will probably start with CS if they have availability between the other work
17:22:21
rucknium:
Code auditing plan sounds good to me. I am no code auditing expert of course :)
17:22:30
rucknium:
Anything else on this agenda item?
17:24:08
jberman:
Nothing from me :)
17:24:14
rucknium:
4. FCMP alpha stressnet (https://monero.town/post/6763165).
17:25:45
jberman:
It seems @ofrnxmr:monero.social has identified a new issue to work through when blocks increase in size that I'll look into today. Tx relay v2 appears to be running smoothly
17:27:14
jberman:
We're still going on beta tasks (scaling, finalizing code, awaiting kayaba's availability to open up as well)
17:30:18
rucknium:
Sounds good. Anything else about stressnet?
17:30:45
jberman:
Nothing from me
17:32:03
rucknium:
We can end the meeting here. Thanks everyone.
17:33:26
articmine:
Thanks
17:33:50
jberman:
thank you
17:43:42
syntheticbird:
A short meeting that's rare
17:43:44
syntheticbird:
thx