19:00:23 <wumpus> #startmeeting
19:00:23 <lightningbot> Meeting started Thu Jan 25 19:00:23 2018 UTC.  The chair is wumpus. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:23 <lightningbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
19:00:42 <provoostenator> Hi
19:00:51 <wumpus> #bitcoin-core-dev Meeting: wumpus sipa gmaxwell jonasschnelli morcos luke-jr btcdrak sdaftuar jtimon cfields petertodd kanzure bluematt instagibbs phantomcircuit codeshark michagogo marcofalke paveljanik NicolasDorier jl2012 achow101 meshcollider jnewbery maaku fanquake promag provoostenator
19:00:55 <jonasschnelli> hi
19:00:55 <achow101> can we release 0.16 yet?
19:01:11 <meshcollider> Hi
19:01:23 <wumpus> today or tomorrow, hopefully
19:01:25 <instagibbs> perhaps an RC first
19:01:28 <wumpus> there's still two PRs open
19:01:32 <wumpus> instagibbs: always
19:01:36 <cfields> hi
19:01:40 <kanzure> hi.
19:01:44 <instagibbs> wumpus, thatsthejoke.gif
19:01:52 <wumpus> https://github.com/bitcoin/bitcoin/milestone/30
19:02:25 <instagibbs> not sure how to evaluated the chain stats one
19:02:26 <jtimon> before releasing we would need to fork from master, no?
19:02:40 <instagibbs> jtimon, can happen immediately after, afaik
19:02:46 <wumpus> so the chaintxdata needs to be checked and BlueMatt's pr reviewed
19:02:48 <instagibbs> well, version bump ofc
19:02:54 <wumpus> jtimon: yes, no way I'm going to forget that
19:02:57 <jtimon> instagibbs: without rc?
19:02:58 <Chris_Stewart_5> hello
19:03:02 <achow101> jtimon: branching happens right before rc is tagged
19:03:11 <jtimon> achow101: right
19:03:21 <wumpus> branch -> [on 0.16 branch] change version information -> tag rc
19:03:29 <cfields> oh shit, I never PR'd the rpc fd exhaustion fix. wumpus: ok to do that right after meeting and add to 0.16 milestone?
19:03:38 <wumpus> -> [on master] change version information -> clear out release notes
19:04:10 <wumpus> 0.16 branch will be 0.16.0 (with release=true) master will be 0.16.99
19:04:13 <wumpus> cfields: sure
19:04:23 <cfields> thanks
19:05:11 <wumpus> though maybe it can wait for 0.16.1 too, I don't know, it's only a local issue
19:06:14 <MarcoFalke> It's a major release. There is no way we will have only one rc
19:06:23 <MarcoFalke> Backport works as well
19:06:23 <jtimon> although not a huge bug, https://github.com/bitcoin/bitcoin/pull/12172 is a bugfix (and a simple one)...
19:06:27 <wumpus> right
19:06:46 <wumpus> please, we want to do the release, we're not going too add more stuff to 0.16.0 now unless absolutely critical
19:07:11 <jonasschnelli> yes... we did already last week... this week is release week. :)
19:07:24 <wumpus> it's the same story every week
19:07:30 <jonasschnelli> heh...
19:07:46 <wumpus> and as MarcoFalke says there's no way there's only going to be one rc
19:07:59 <wumpus> there are going to be problems that come up in rc1 and have to be fixed
19:08:23 <jtimon> it's not like that +9-4 change is going to break anything, but ok
19:08:34 <MarcoFalke> and release notes to be written ...
19:08:56 <MarcoFalke> jtimon: It still needs review. I told you last week
19:09:20 <jtimon> MarcoFalke: sure, it needs more review
19:09:59 <jtimon> anyway, never mind
19:10:05 <MarcoFalke> jtimon: I like it, but others need to like it as well
19:10:28 <meshcollider> Release notes for major releases are so much longer than point releases but I'll try and find time if noone else does
19:10:58 <MarcoFalke> meshcollider: Even doing half of it would help
19:10:58 <wumpus> maybe we could put it in the wiki again?
19:11:06 <MarcoFalke> oh right
19:11:13 <meshcollider> wumpus: yeah +1
19:11:25 <wumpus> #action put release 0.16 notes in wiki
19:11:27 <achow101> yes, let's do the wiki thing again
19:12:17 <BlueMatt> yes, #12266 needs review :/
19:12:18 <wumpus> and re: #12270 does anyone know the exact process there?
19:12:18 <gribble> https://github.com/bitcoin/bitcoin/issues/12266 | Move scheduler/threadGroup into common-init instead of per-app by TheBlueMatt · Pull Request #12266 · bitcoin/bitcoin · GitHub
19:12:20 <gribble> https://github.com/bitcoin/bitcoin/issues/12270 | Update chainTxData for 0.16 by laanwj · Pull Request #12270 · bitcoin/bitcoin · GitHub
19:12:22 <MarcoFalke> https://github.com/bitcoin-core/bitcoin-devwiki/wiki/0.16.0-Release-notes
19:12:43 <wumpus> the release notes are quite unclear on chainTxData
19:12:47 <BlueMatt> wumpus: maybe sipa can confirm, he's currently mid-giving-talk
19:12:59 <kanzure> git clone https://github.com/bitcoin-core/bitcoin-devwiki.wiki.git
19:13:03 <cfields> BlueMatt: is there q/a after?
19:13:09 <BlueMatt> lol
19:13:55 <kanzure> oh bpase scheduling
19:14:06 <cfields> BlueMatt: thanks for the fix, btw. taking a look.
19:14:21 <meshcollider> Are bpase talks recorded btw
19:14:28 <kanzure> site says videos will be posted later
19:15:49 <wumpus> any other topics?
19:16:12 <achow101> coin selection?
19:16:19 <achow101> (again)
19:16:31 <wumpus> #topic coin selection
19:16:40 <Murch> Currently listening to sipa @ BPASE :p
19:16:59 <achow101> I did some more simulations, this time with varying fee rates
19:17:02 <achow101> results here: https://github.com/bitcoin/bitcoin/pull/10637#issuecomment-359514676
19:17:07 <wumpus> I saw a new PR was openened about that #12257, haven't had time to look at it yet
19:17:09 <gribble> https://github.com/bitcoin/bitcoin/issues/12257 | [wallet] Use destination groups instead of coins in coin select by kallewoof · Pull Request #12257 · bitcoin/bitcoin · GitHub
19:17:46 <achow101> the results don't seem to be very impressive though
19:18:00 <meshcollider> wumpus: that's pretty orthogonal I think, it's a privacy thing rather than an efficient selection thing
19:18:12 <wumpus> ok...
19:18:58 <achow101> I'm not sure of BnB is going to be much of an improvement over what we currently have
19:19:15 <achow101> s/of/that/
19:19:35 <achow101> at the very least, it doesn't do worse
19:19:49 <wumpus> that's already good if it is a win in runtime
19:20:09 <achow101> it actually runs a lot slower
19:20:15 <meshcollider> Oh
19:20:21 <wumpus> ouch
19:20:39 <achow101> I noticed during the simulations that running BnB takes longer since it has to exhaust the search before it continues with the current algo
19:20:46 <morcos> Should we save this conversation for when Murch can join?
19:21:04 <achow101> It seems to only be faster when it can find a solution, which is not all that frequent
19:21:59 <Murch> achow101: I just had a look at the results. How come that there were several hundred instances of BnB use but only one fewer change output created than without BnB?
19:22:47 <Murch> I think there is an unexpected behavior there.
19:23:07 <achow101> oh, that's strange
19:26:08 <achow101> I will look into that
19:27:21 <wumpus> thanks
19:27:22 <wumpus> any other topics?
19:27:36 <Chris_Stewart_5> achow101: It seems like the algo would obviously be slower if it reverts to the old algo?
19:27:45 <jtimon> perhaps a short meeting today?
19:27:46 <achow101> Chris_Stewart_5: yes
19:28:01 <meshcollider> jtimon: yes seems sipa's talk has stolen the show today :)
19:28:03 <achow101> Chris_Stewart_5: logically, reverting means that it's running more code, so it kinda has to be slower
19:28:04 <Chris_Stewart_5> so the high failure rate is the main concern?
19:28:20 <jtimon> meshcollider: oh, what talk?
19:28:39 <meshcollider> jtimon: at bpase
19:28:40 <wumpus> jtimon: yep
19:28:54 <wumpus> let's review stuff for 0.16 after the talk :)
19:29:55 <jtimon> oh, I'll wait for the video
19:30:28 <wumpus> #endmeeting