--- Log opened Wed Jun 04 00:00:04 2025
00:20 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
00:58 -!- l0rinc [~l0rinc@user/l0rinc] has joined #bitcoin-core-pr-reviews
01:24 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 268 seconds]
02:08 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
02:13 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 276 seconds]
02:39 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
03:00 -!- Netsplit *.net <-> *.split quits: greypw1495085720, ghost43
03:01 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 252 seconds]
03:29 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
03:45 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 260 seconds]
03:59 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
04:16 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 252 seconds]
04:57 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
05:01 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 252 seconds]
05:30 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
05:46 -!- kevkevin_ [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
05:48 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 260 seconds]
05:53 -!- kevkevin_ [~kevkevin@209.242.39.30] has quit [Ping timeout: 276 seconds]
06:03 -!- l0rinc [~l0rinc@user/l0rinc] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
06:12 -!- l0rinc [~l0rinc@user/l0rinc] has joined #bitcoin-core-pr-reviews
06:14 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
06:58 -!- l0rinc [~l0rinc@user/l0rinc] has quit [Read error: Connection reset by peer]
07:02 -!- jonatack [~jonatack@user/jonatack] has quit [Ping timeout: 272 seconds]
07:05 -!- jonatack [~jonatack@user/jonatack] has joined #bitcoin-core-pr-reviews
07:32 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
07:38 -!- robszarka [~szarka@2603:3003:4eac:100:1caf:9753:ca8c:3cbd] has joined #bitcoin-core-pr-reviews
07:41 -!- szarka [~szarka@2603:3003:4eac:100:d555:3e6d:b41c:f882] has quit [Ping timeout: 248 seconds]
08:05 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Quit: grettke]
08:38 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
09:53 -!- l0rinc [~l0rinc@user/l0rinc] has joined #bitcoin-core-pr-reviews
10:00 < marcofleon> #startmeeting
10:00 < corebot> marcofleon: Meeting started at 2025-06-04T17:00+0000
10:00 < corebot> marcofleon: Current chairs: marcofleon
10:00 < corebot> marcofleon: Useful commands: #action #info #idea #link #topic #motion #vote #close #endmeeting
10:00 < corebot> marcofleon: See also: https://hcoop-meetbot.readthedocs.io/en/stable/
10:00 < corebot> marcofleon: Participants should now identify themselves with '#here' or with an alias like '#here FirstLast'
10:01 < sipa> hi
10:02 < marcofleon> sipa did you review the pr y/n
10:02 < sipa> errrr
10:02 < abubakarsadiq> hi
10:02 < sipa> i may have glanced over it
10:03 < abubakarsadiq> haha
10:03 < marcofleon> lesson learned. don't put up notes late. At least I really reviewed it well by doing this
10:04 < sipa> Concept ACK, approach ACK, tested ACK, or NACK? What was your review approach?
10:04 < marcofleon> Concept ACK and have fuzzed the targets a bit as well
10:05 < glozow> hi
10:05 < abubakarsadiq> haven't fuzz the new update yet. Did not take a look at the changes after @marcofleon crash report.
10:06 < marcofleon> I feel the new targets that test the helpers are fairly straightforwward. It's more of the clusterlin stuff in general that is tricky
10:06 < marcofleon> abubakarsadiq no worries, it was a very simple change to address that crash
10:10 < abubakarsadiq> yep
10:10 < marcofleon> I will say, going over this may have inspired me to try to tackle the SFL PR
10:11 < sipa> please do!
10:12 < abubakarsadiq> Yeah but remarkably excellent docs to guide the clusterlin learnings.
10:14 < marcofleon> Thanks! I don't think going over the questions would be too valuable for this particular group so i'll end meeting
10:14 < glozow> marcofleon: 🚀
10:14 < marcofleon> #endmeeting
10:14 < corebot> marcofleon: Meeting ended at 2025-06-04T17:14+0000
10:14 < corebot> marcofleon: Raw log: https://achow101.com/ircmeetings/2025/bitcoin-core-pr-reviews.2025-06-04_17_00.log.json
10:14 < corebot> marcofleon: Formatted log: https://achow101.com/ircmeetings/2025/bitcoin-core-pr-reviews.2025-06-04_17_00.log.html
10:14 < corebot> marcofleon: Minutes: https://achow101.com/ircmeetings/2025/bitcoin-core-pr-reviews.2025-06-04_17_00.html
10:14 < abubakarsadiq> thanks for writing the notes
10:14 < glozow> oh!
10:14 < sipa> Still happy to chat about the PR, or related stuff :)
10:15 < glozow> thank you marcofleon
10:16 < abubakarsadiq> >  no worries, it was a very simple change to address that crash
10:16 < abubakarsadiq> yep haven't seen the range diff after all the rebase. Do you have that noted somewhere
10:23 < marcofleon> sipa: I haven't looked much at the sfl stuff, would need to read the delving. But ancestor finder in the fuzz test goes away because the structure changes yes? The tree part of the algorithm takes care of that
10:24 < sipa> marcofleon: not quite
10:25 < sipa> it goes away because (a) there is no good way to incorporate it, and (b) it's likely not really a gain anyway, making it not that important to go out of our way to incorporate it still
10:26 < glozow> marcofleon: abubakarsadiq: if you want to redo the meeting later this week after more review time, we could do that?
10:26 < sipa> structure, the existing linearization algorithm at a high-level is LIMO, which at a very high level boils down to: start with an existing linearization, and then find good high-feerate toplogically-valid sets to move to the front, in a way that never worsens the existing the linearization
10:27 < sipa> to find these sets to move the front, it uses ancestor-set-search (O(N^2)) and candidate-set-search (O(N*sqrt(2^N), but stopping when running out of time)
10:28 < sipa> it uses both, because ancestor-set-search is fast and okayish (sufficient for CPFP), rather than just candidate-set-search which may spend a whole lot of time and not find anything
10:28 < sipa> now for SFL
10:29 -!- gophern [~gophern@105.113.114.245] has joined #bitcoin-core-pr-reviews
10:29 < sipa> it doesn't use LIMO's structure, of operating on an existing linearization that is being improved, so there isn't a place for putting arbitrary "find good subsets" as a means to improve it
10:30 < sipa> instead, it maintains a set of trees that span the graph, and keeps modifying these - there is a way to convert an existing linearization to an SFL state, and a way to convert an SFL to a linearization, in a way that still has the "always better than what you started from" property
10:30 < marcofleon> abubakarsadiq: I think it was just a matter of changing ClusterIndex to DepgraphIndex? I'm not entirely sure, I'll check it out again
10:31 < marcofleon> glozow: I could be down for that. I was gonna post my review in the next couple of days so that could help
10:31 < sipa> but other than in LIMO, the state is more than just a linearization, and there isn't really a way to make it do good-subset-driven improvements to it, the improvements are found directly from the structure
10:32 < abubakarsadiq> glozow: I am up for it
10:33 < sipa> and for (b)... my guess is that per amount of time spent, SFL makes better decisions than ancestor set search, so even if there was a way to integrate it, it's probably not a win
10:33 < sipa> a downside is that this means we don't get a "guaranteed CPFP quality" feeling anymore (i say feeling, because we've tried and failed to actually define what properties such a quality has)
10:34 < sipa> the upside is it's just so much faster that i think in practice nearly all non-pathological clusters will just be sorted optimally
10:34 < sipa> does that answer things a bit?
10:36 < sipa> maybe a more direct answer: the tests for ancestor finding go away, because ancestor finding goes away
10:36 < abubakarsadiq> > because we've tried and failed to actually define what properties such a quality
10:36 < abubakarsadiq> Improved fee rate diagram no?
10:36 < abubakarsadiq> > a downside is that this means we don't get a "guaranteed CPFP quality" feeling anymore
10:36 < abubakarsadiq> how is that not bad though; 
10:37 < sipa> abubakarsadiq: we can define "linearization A is better than linearization B" just find
10:37 < sipa> but we can't define what specific quality properties a linearization has that is constructed by incorporating ancestor finding
10:38 < sipa> it feels like it ought to be "enough for CPFP", but in complicated settings, this isn't actually true
10:38 < sipa> and in simple settings... well those are so trivial anyway they'll be optimal
10:44 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Quit: grettke]
10:46 < marcofleon> sipa: that did help thank you. Time to.. delve into bitcoin
10:49 -!- gophern [~gophern@105.113.114.245] has quit [Quit: Client closed]
10:49 -!- gophern [~gophern@105.113.114.245] has joined #bitcoin-core-pr-reviews
10:52 -!- gophern [~gophern@105.113.114.245] has quit [Client Quit]
10:54 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
10:58 < sipa> badum tss
11:01 -!- Guest15 [~Guest81@host-79-42-195-146.retail.telecomitalia.it] has joined #bitcoin-core-pr-reviews
11:01 -!- Guest15 [~Guest81@host-79-42-195-146.retail.telecomitalia.it] has quit [Client Quit]
11:16 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has joined #bitcoin-core-pr-reviews
11:22 -!- l0rinc [~l0rinc@user/l0rinc] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
12:01 -!- l0rinc [~l0rinc@user/l0rinc] has joined #bitcoin-core-pr-reviews
12:26 -!- robszarka [~szarka@2603:3003:4eac:100:1caf:9753:ca8c:3cbd] has quit [Quit: Leaving]
12:26 -!- szarka [~szarka@2603:3003:4eac:100:1caf:9753:ca8c:3cbd] has joined #bitcoin-core-pr-reviews
12:27 < sliv3r__> Noooo I missed this
12:28 < sliv3r__> glozow: marcofleon: abubakarsadiq if you redo the meeting next week to have time to review ping here please :)
12:37 -!- Guest2 [~Guest88@105.163.158.178] has joined #bitcoin-core-pr-reviews
12:38 -!- Guest2 [~Guest88@105.163.158.178] has quit [Client Quit]
12:38 -!- devgitotox [~devgitoto@105.163.158.178] has joined #bitcoin-core-pr-reviews
12:38 -!- devgitotox [~devgitoto@105.163.158.178] has quit [Client Quit]
12:40 -!- thelounge49 [~thelounge@149.106.235.56] has quit [Quit: The Lounge - https://thelounge.chat]
12:42 -!- thelounge49 [~thelounge@149.106.235.56] has joined #bitcoin-core-pr-reviews
12:59 -!- l0rinc [~l0rinc@user/l0rinc] has quit [Ping timeout: 272 seconds]
13:26 -!- thelounge49 [~thelounge@149.106.235.56] has quit [Quit: The Lounge - https://thelounge.chat]
13:27 -!- thelounge49 [~thelounge@149.106.235.56] has joined #bitcoin-core-pr-reviews
13:30 -!- jamesob156659 [~jamesob@pool-108-44-244-6.clppva.fios.verizon.net] has quit [Ping timeout: 260 seconds]
13:30 -!- jamesob443688173 [~jamesob@pool-108-44-244-6.clppva.fios.verizon.net] has quit [Ping timeout: 260 seconds]
13:48 -!- jamesob156659 [~jamesob@pool-108-44-244-6.clppva.fios.verizon.net] has joined #bitcoin-core-pr-reviews
13:48 -!- jamesob443688173 [~jamesob@pool-108-44-244-6.clppva.fios.verizon.net] has joined #bitcoin-core-pr-reviews
14:38 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Quit: grettke]
15:24 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
15:59 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Quit: grettke]
16:54 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
18:00 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Quit: grettke]
18:42 -!- grettke [~grettke@184.55.133.0] has joined #bitcoin-core-pr-reviews
19:24 -!- grettke [~grettke@184.55.133.0] has quit [Quit: grettke]
20:32 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
20:33 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Client Quit]
20:45 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has joined #bitcoin-core-pr-reviews
21:06 -!- grettke [~grettke@syn-184-055-133-000.res.spectrum.com] has quit [Quit: grettke]
21:34 -!- abubakarsadiq [uid602234@id-602234.hampstead.irccloud.com] has quit [Quit: Connection closed for inactivity]
22:35 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Remote host closed the connection]
23:06 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
23:11 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 248 seconds]
23:26 -!- kevkevin [~kevkevin@209.242.39.30] has joined #bitcoin-core-pr-reviews
23:38 -!- kevkevin [~kevkevin@209.242.39.30] has quit [Ping timeout: 268 seconds]
--- Log closed Thu Jun 05 00:00:06 2025