12017-01-17T00:03:49  *** laurentmt has joined #bitcoin-core-dev
  22017-01-17T00:05:13  *** laurentmt has quit IRC
  32017-01-17T00:09:49  *** Chris_Stewart_5 has joined #bitcoin-core-dev
  42017-01-17T00:29:36  *** kvnn has quit IRC
  52017-01-17T00:41:36  *** Chris_Stewart_5 has quit IRC
  62017-01-17T00:47:29  *** Ylbam has quit IRC
  72017-01-17T00:50:53  *** abpa has joined #bitcoin-core-dev
  82017-01-17T00:50:58  *** abpa has quit IRC
  92017-01-17T01:06:43  *** MarcoFalke has joined #bitcoin-core-dev
 102017-01-17T01:16:36  *** Netmage has quit IRC
 112017-01-17T01:17:36  *** Netmage has joined #bitcoin-core-dev
 122017-01-17T01:44:54  *** MarcoFalke has quit IRC
 132017-01-17T02:00:30  *** Guest38975 has joined #bitcoin-core-dev
 142017-01-17T02:07:06  *** Alopex has quit IRC
 152017-01-17T02:08:11  *** Alopex has joined #bitcoin-core-dev
 162017-01-17T02:12:51  *** jtimon has quit IRC
 172017-01-17T02:19:37  <BlueMatt> wumpus: fwiw, i vote we hold merge window for a day or two for things already pending incl the hd split (which I think is one more set of fixups, hopefully tomorrow, away, and is not as hard to review as it looks), bumpfee pending discussion tomorrow (I'm pretty happy with it if we fix the two pending issues - listunspent, which may just be a docs change, and getbalance, which may also just be a docs change), the compact-block-orphan-r
 182017-01-17T02:19:37  <BlueMatt> ejects one (which looks pretty much mergeable to me) and preferably the extra message handler waker (9561, which is both trivial and a big gain)
 192017-01-17T02:19:47  <BlueMatt> but I'd understand if thats not a popular opinion
 202017-01-17T02:29:22  *** fanquake has joined #bitcoin-core-dev
 212017-01-17T02:29:31  <fanquake> wummpus / sipa are you around?
 222017-01-17T02:29:35  <fanquake> *wumpus
 232017-01-17T02:34:21  <fanquake> Or anyone with admin on bitcoin/bitcoin, there is a user spamming the repo with links/comments.
 242017-01-17T02:35:34  <fanquake> I have deleted most for now.
 252017-01-17T02:43:20  <sipa>  sigh
 262017-01-17T02:43:28  <sipa> sorry, i can't really access github now
 272017-01-17T02:53:09  *** dermoth_ has joined #bitcoin-core-dev
 282017-01-17T02:53:37  *** dermoth has quit IRC
 292017-01-17T02:53:39  *** dermoth_ is now known as dermoth
 302017-01-17T03:25:40  <sipa> fanquake: blocked
 312017-01-17T03:55:15  <bitcoin-git> [bitcoin] sipa pushed 3 new commits to master: https://github.com/bitcoin/bitcoin/compare/812714fd80e9...6696b4635ceb
 322017-01-17T03:55:16  <bitcoin-git> bitcoin/master f13914a Matt Corallo: Make WakeMessageHandler public
 332017-01-17T03:55:16  <bitcoin-git> bitcoin/master 241d893 Matt Corallo: Wake message handling thread when we receive a new block...
 342017-01-17T03:55:17  <bitcoin-git> bitcoin/master 6696b46 Pieter Wuille: Merge #9561: Wake message handling thread when we receive a new block...
 352017-01-17T03:55:30  <bitcoin-git> [bitcoin] sipa closed pull request #9561: Wake message handling thread when we receive a new block (master...2017-01-wakeup-on-new-block) https://github.com/bitcoin/bitcoin/pull/9561
 362017-01-17T04:05:13  *** CubicEarth has joined #bitcoin-core-dev
 372017-01-17T04:12:39  <sipa> BlueMatt: sorry, need sleep first
 382017-01-17T04:13:14  <BlueMatt> I still vote we push back freeze a day or two, given today was a holiday in the us and many folks werent working all weekend
 392017-01-17T04:13:18  <BlueMatt> :p
 402017-01-17T04:14:12  <BlueMatt> and so I can shamelessly get folks to review #9535 :p
 412017-01-17T04:14:13  <gribble> https://github.com/bitcoin/bitcoin/issues/9535 | Split CNode::cs_vSend: message processing and message sending by TheBlueMatt · Pull Request #9535 · bitcoin/bitcoin · GitHub
 422017-01-17T04:31:31  <luke-jr> I ended up being away for yesterday & today, so another day or two may be helpful
 432017-01-17T04:31:55  *** Victor_sueca has joined #bitcoin-core-dev
 442017-01-17T04:31:59  *** CubicEarth has quit IRC
 452017-01-17T04:34:35  *** Victorsueca has quit IRC
 462017-01-17T04:48:59  *** CubicEarth has joined #bitcoin-core-dev
 472017-01-17T05:52:08  *** Squidicc has joined #bitcoin-core-dev
 482017-01-17T05:54:28  *** Squidicuz has quit IRC
 492017-01-17T06:10:54  *** fanquake has quit IRC
 502017-01-17T07:01:51  *** Ylbam has joined #bitcoin-core-dev
 512017-01-17T07:12:43  *** waxwing has quit IRC
 522017-01-17T08:10:29  *** BashCo has quit IRC
 532017-01-17T08:11:13  *** BashCo has joined #bitcoin-core-dev
 542017-01-17T08:15:31  *** BashCo has quit IRC
 552017-01-17T08:23:45  *** Guyver2 has joined #bitcoin-core-dev
 562017-01-17T08:37:28  *** BashCo has joined #bitcoin-core-dev
 572017-01-17T08:43:26  *** CubicEarth has quit IRC
 582017-01-17T08:47:35  * jonasschnelli request a retest of: https://github.com/bitcoin/bitcoin/pull/9294 from BlueMatt luke-jr (thanks in advance!)
 592017-01-17T09:18:59  *** wasi has quit IRC
 602017-01-17T09:21:18  *** paveljanik has quit IRC
 612017-01-17T09:24:46  <wumpus> I've extended the section on named arguments in the release notes a bit: https://github.com/bitcoin-core/bitcoin-devwiki/wiki/0.14.0-Release-notes#support-for-json-rpc-named-arguments
 622017-01-17T09:25:24  *** wasi has joined #bitcoin-core-dev
 632017-01-17T09:25:33  *** jtimon has joined #bitcoin-core-dev
 642017-01-17T09:54:08  *** Guyver2 has quit IRC
 652017-01-17T10:18:27  *** droark has quit IRC
 662017-01-17T10:23:59  *** waxwing__ has joined #bitcoin-core-dev
 672017-01-17T10:25:16  *** justanotheruser has quit IRC
 682017-01-17T10:32:03  *** wvr has joined #bitcoin-core-dev
 692017-01-17T10:37:01  *** waxwing__ is now known as waxwing
 702017-01-17T10:39:40  *** BashCo has quit IRC
 712017-01-17T10:40:02  *** fanquake has joined #bitcoin-core-dev
 722017-01-17T10:43:08  *** waxwing has quit IRC
 732017-01-17T10:46:46  *** MarcoFalke has joined #bitcoin-core-dev
 742017-01-17T10:47:18  <bitcoin-git> [bitcoin] jdust69 opened pull request #9568: 10.13 (master...0.13) https://github.com/bitcoin/bitcoin/pull/9568
 752017-01-17T10:48:14  <bitcoin-git> [bitcoin] fanquake closed pull request #9568: 10.13 (master...0.13) https://github.com/bitcoin/bitcoin/pull/9568
 762017-01-17T10:49:05  *** aalex has quit IRC
 772017-01-17T10:49:23  *** BashCo has joined #bitcoin-core-dev
 782017-01-17T10:49:26  *** aalex has joined #bitcoin-core-dev
 792017-01-17T10:56:08  *** waxwing has joined #bitcoin-core-dev
 802017-01-17T11:10:50  *** waxwing has quit IRC
 812017-01-17T11:15:10  *** MarcoFalke has quit IRC
 822017-01-17T11:21:53  *** BashCo has quit IRC
 832017-01-17T11:21:59  *** BashCo_ has joined #bitcoin-core-dev
 842017-01-17T11:23:44  *** waxwing has joined #bitcoin-core-dev
 852017-01-17T12:16:29  <bitcoin-git> [bitcoin] fanquake closed pull request #9263: release notes: Explicitly mention the removal of free transactions, and do not commit to removal of priority in any given release (master...relnotes_freetxn) https://github.com/bitcoin/bitcoin/pull/9263
 862017-01-17T12:31:03  *** jannes has joined #bitcoin-core-dev
 872017-01-17T12:34:31  *** laurentmt has joined #bitcoin-core-dev
 882017-01-17T12:35:46  *** Victor_sueca has quit IRC
 892017-01-17T12:36:08  *** laurentmt has quit IRC
 902017-01-17T12:36:29  <morcos> luke-jr: what is the point of doing the release notes on a wiki, if you are going to just merge changes that were explicitly NACK'ed when those same changs were proposed in a PR
 912017-01-17T12:36:48  <morcos> all that is going to do is lead to a childish edit war and make it so none of us can use the wiki for release notes
 922017-01-17T12:37:05  <morcos> if you know you have a controversial change you want to make you shouldln't just make it and hope no one notices
 932017-01-17T12:37:19  <morcos> this is not a proper way of working with other people
 942017-01-17T12:59:29  *** cryptapus has joined #bitcoin-core-dev
 952017-01-17T13:03:59  *** fanquake has quit IRC
 962017-01-17T13:14:21  *** laurentmt has joined #bitcoin-core-dev
 972017-01-17T13:15:35  *** waxwing has quit IRC
 982017-01-17T13:25:00  *** justanotheruser has joined #bitcoin-core-dev
 992017-01-17T13:30:37  *** laurentmt has quit IRC
1002017-01-17T13:32:17  *** waxwing has joined #bitcoin-core-dev
1012017-01-17T14:07:15  <BlueMatt> jonasschnelli: hey
1022017-01-17T14:07:22  <BlueMatt> how much longer will you be around?
1032017-01-17T14:07:31  <jonasschnelli> BlueMatt: 1-2h at least
1042017-01-17T14:07:34  <jonasschnelli> hi
1052017-01-17T14:07:46  <BlueMatt> ok, will circle back around after breakfast and hopefully we can finish this
1062017-01-17T14:08:09  <jonasschnelli> Yes. I guess and hope I fixed most/all of you points... tell me, if there is more
1072017-01-17T14:08:31  <BlueMatt> i think it should be good if you hit all the stuff i commented on, i gave it a pretty decent review the first go-around, but it needs another one when I'm awake post-breakfast
1082017-01-17T14:08:54  <jonasschnelli> Yes. I think your review was very valuable...
1092017-01-17T14:09:14  <jonasschnelli> We should not rush that change... but hurry up for 0.14. :)=
1102017-01-17T14:10:17  <BlueMatt> heh, yea, I recommended pushing back yesterday because I'm confident we can get this one in
1112017-01-17T14:10:46  <jonasschnelli> Yes. I mean if we find something after the freeze, there is time to fix it before the 0.14 release...
1122017-01-17T14:11:06  <BlueMatt> yup
1132017-01-17T14:19:00  <sipa> do we want to be a mentoring org for GSOC?
1142017-01-17T14:20:19  <BlueMatt> does anyone feel like they have time?
1152017-01-17T14:20:44  <BlueMatt> i mean in theory, maybe, but....
1162017-01-17T14:20:51  <sipa> yeah...
1172017-01-17T14:21:25  <sipa> i feel like we should make time for such projects
1182017-01-17T14:21:36  <sipa> but easier said than done
1192017-01-17T14:22:01  <BlueMatt> yea
1202017-01-17T14:22:08  <BlueMatt> obviously agreed
1212017-01-17T14:25:04  *** laurentmt has joined #bitcoin-core-dev
1222017-01-17T14:29:24  *** laurentmt has quit IRC
1232017-01-17T14:44:25  <BlueMatt> jonasschnelli: any chance we can s/(used for change outputs, only appears if HD is enabled otherwise there is no need for internal keys)/(used for change outputs, only appears if HD split is enabled, otherwise external keys are used)/
1242017-01-17T14:44:47  <BlueMatt> and make the corresponding change the the logic
1252017-01-17T14:44:53  <BlueMatt> should make the docs wayyy clearer
1262017-01-17T14:45:09  <jonasschnelli> BlueMatt: Okay. Makese sense I guess.
1272017-01-17T14:45:15  <jonasschnelli> I liked the idea form luke-jr...
1282017-01-17T14:45:28  <jonasschnelli> keypoolsize: {"internal": x, "external": y}
1292017-01-17T14:45:31  <jonasschnelli> But breaks the API
1302017-01-17T14:45:39  <BlueMatt> yea, agreed lets not break the api
1312017-01-17T14:45:46  <jonasschnelli> okay.. will update
1322017-01-17T14:46:26  <jonasschnelli> BlueMatt. one meh though...
1332017-01-17T14:46:36  <jonasschnelli> Do we want to expose "HD split is enabled" to the user?
1342017-01-17T14:46:52  <jonasschnelli> "HD split" as a label can be confusing...
1352017-01-17T14:47:05  <BlueMatt> hum, or maybe just "only appears if wallet is using this feature, otherwise external keys are used"?
1362017-01-17T14:47:11  <jonasschnelli> We could name it "HD internal keypool" or similar
1372017-01-17T14:47:19  <jonasschnelli> yes. Your text is good
1382017-01-17T14:47:23  <jonasschnelli> Let me take it
1392017-01-17T14:47:28  <BlueMatt> ok, cool
1402017-01-17T14:51:23  <jonasschnelli> BlueMatt: https://github.com/bitcoin/bitcoin/pull/9294/commits/eeeb52afc3b25836588b2b7c6b704a5a6498e1d1
1412017-01-17T14:51:30  <jonasschnelli> Added to #9294
1422017-01-17T14:51:33  <gribble> https://github.com/bitcoin/bitcoin/issues/9294 | Use internal HD chain for change outputs (hd split) by jonasschnelli · Pull Request #9294 · bitcoin/bitcoin · GitHub
1432017-01-17T14:51:43  <BlueMatt> thanks
1442017-01-17T14:58:03  *** jnewbery1 has joined #bitcoin-core-dev
1452017-01-17T15:01:18  <BlueMatt> jonasschnelli: do we normally bump wallet versions at release? I think we just set them based on next version pre-merge
1462017-01-17T15:01:49  <jonasschnelli> BlueMatt: in 0.13, we bumped to 130000 after the tag/version.bump was made
1472017-01-17T15:02:08  <BlueMatt> ehh, ok
1482017-01-17T15:02:19  <jonasschnelli> Otherwise you can't test it and all tests will fail
1492017-01-17T15:02:27  <BlueMatt> oh, ok, didnt realize that
1502017-01-17T15:02:35  <BlueMatt> can you not catch(...) in deserialize?
1512017-01-17T15:02:45  <BlueMatt> we always throw std::ios_base::failure when we read oob
1522017-01-17T15:02:57  <jonasschnelli> BlueMatt: the problem there is, that CKeyPool has no record version...
1532017-01-17T15:03:06  <jonasschnelli> maybe we could check is CDataStream has more bytes?
1542017-01-17T15:03:10  <BlueMatt> yes, catch(std::ios_base::failure&)
1552017-01-17T15:03:16  <jonasschnelli> ah.. okay.
1562017-01-17T15:03:39  <BlueMatt> otherwise you'll catch all kinds of fun garbage that should really reach top-of-thread and assert
1572017-01-17T15:03:48  <jonasschnelli> Yes. Indeed.
1582017-01-17T15:05:02  *** Chris_Stewart_5 has joined #bitcoin-core-dev
1592017-01-17T15:06:57  *** Sosumi has joined #bitcoin-core-dev
1602017-01-17T15:07:19  <jonasschnelli> BlueMatt: https://github.com/bitcoin/bitcoin/pull/9294/commits/a874b75a180a9e967a3d936cf46df11f8531b70a
1612017-01-17T15:07:46  <jonasschnelli> I wonder if a simple "are there more bytes?" check would be more appropriate at this place
1622017-01-17T15:08:02  *** morph has joined #bitcoin-core-dev
1632017-01-17T15:08:28  <jonasschnelli> On the other hand, a failed deserialization at this point shoud always flag the CKeyPool item fInternal=false
1642017-01-17T15:13:03  <BlueMatt> jonasschnelli: ok, the only remaining concern i have is performance - there are a bunch of loops over disk reads introduced for folks who upgrade
1652017-01-17T15:13:35  <jonasschnelli> Yes. This could be optimised...
1662017-01-17T15:13:51  <jonasschnelli> I expect BDB to be sort of fast...
1672017-01-17T15:14:11  <BlueMatt> I think you may be surprised
1682017-01-17T15:14:15  <BlueMatt> but if its an issue its an easy fix
1692017-01-17T15:14:25  <instagibbs> that can be fixed post-freeze, yeS?
1702017-01-17T15:14:30  <jonasschnelli> Yes. CKeyPool should have been fixed long time ago...
1712017-01-17T15:15:00  <BlueMatt> instagibbs: if it turns out to be a major regression, I'd say yes
1722017-01-17T15:15:19  <jonasschnelli> BlueMatt: I can test it with a 100k keypool wallet... and compare
1732017-01-17T15:16:43  <jonasschnelli> Oh.. lets better start with 50k...
1742017-01-17T15:17:08  <jonasschnelli> or with 5k,... CKey.Verify() is slow..
1752017-01-17T15:17:31  <BlueMatt> heh
1762017-01-17T15:18:03  <sipa> CKey.Verify() can verify 10000 keys per second
1772017-01-17T15:18:15  <sipa> the slow part is the sync to bdb after every keypool change
1782017-01-17T15:22:49  <jonasschnelli> okay.. I see..
1792017-01-17T15:37:25  <jonasschnelli> BlueMatt: With a 5k wallet, its slightly slower on my Core i7 2.9 GHZ..
1802017-01-17T15:37:28  <jonasschnelli> getnewaddress: real	0m0.219s
1812017-01-17T15:39:30  <BlueMatt> vs?
1822017-01-17T15:40:08  <jonasschnelli> real	0m0.104s
1832017-01-17T15:40:11  <jonasschnelli> (0.13)
1842017-01-17T15:41:08  <jonasschnelli> BlueMatt: I think it's acceptable but we should work on a fix/better-memory map
1852017-01-17T15:41:19  <BlueMatt> I think its fine for now, I'll let phantomcircuit fix it when he complains
1862017-01-17T15:42:05  <bitcoin-git> [bitcoin] TheBlueMatt closed pull request #9419: Stop Using cs_main for CNodeState/State() (master...2016-12-nodestate) https://github.com/bitcoin/bitcoin/pull/9419
1872017-01-17T15:43:08  <bitcoin-git> [bitcoin] TheBlueMatt closed pull request #9488: Parallel ThreadMessageHandler (master...2017-01-parallel-processmessages) https://github.com/bitcoin/bitcoin/pull/9488
1882017-01-17T15:43:40  <jonasschnelli> BlueMatt: oh. Keep in mind, that 0.13er wallet has only 5k keys while the 0.14er wallet has 200% key (=10k)
1892017-01-17T15:43:40  <jonasschnelli> So.. then the number look even better
1902017-01-17T16:01:00  *** laurentmt has joined #bitcoin-core-dev
1912017-01-17T16:10:06  <morph> hi
1922017-01-17T16:15:22  *** Victorsueca has joined #bitcoin-core-dev
1932017-01-17T16:26:11  *** Victorsueca has quit IRC
1942017-01-17T16:31:12  *** Victorsueca has joined #bitcoin-core-dev
1952017-01-17T16:51:36  *** abpa has joined #bitcoin-core-dev
1962017-01-17T16:57:33  *** laurentmt has quit IRC
1972017-01-17T17:01:13  *** laurentmt has joined #bitcoin-core-dev
1982017-01-17T17:04:20  *** BashCo_ has quit IRC
1992017-01-17T17:12:51  <luke-jr> morcos: YOUR changes were NACK'd.
2002017-01-17T17:15:01  *** Victor_sueca has joined #bitcoin-core-dev
2012017-01-17T17:15:34  *** Victorsueca has quit IRC
2022017-01-17T17:16:11  <morcos> luke-jr: i didn't have any release note changes except 9531 which were merged with only ACK's. But I'm not going to waste everyone's time arguing with you about this.  I find you very difficult to work with, and so I will just avoid it where possible.
2032017-01-17T17:16:53  <morcos> I've made my view point on the release notes clear.. If you want to go around changing things behind people's backs, I'm not going to take on the task of trying to stop you.  But you should be aware your actions reflect poorly on all of us.
2042017-01-17T17:17:35  <luke-jr> morcos: trying to use Core as a vehicle for political agendas is what reflects poorly.
2052017-01-17T17:18:31  <sipa> luke-jr: stop it
2062017-01-17T17:18:43  <sipa> there have been more than enough arguments
2072017-01-17T17:18:50  <sipa> sorry, people disagree with you
2082017-01-17T17:18:53  <sipa> accept it
2092017-01-17T17:19:19  <luke-jr> disagreement means we should agree to disagree by having an option, not that you should just get your way and force your opinion on others
2102017-01-17T17:19:30  <luke-jr> this would all be over if you would just leave it be.
2112017-01-17T17:20:22  <BlueMatt> this may be true for some thing, but we dont generally support options that are used by +/- one person, especially if they have really big costs to doing so
2122017-01-17T17:20:34  <BlueMatt> s
2132017-01-17T17:20:47  <luke-jr> BlueMatt: the cost to leaving the code in place until it gets in someone's de facto way is literally zero.
2142017-01-17T17:21:21  <BlueMatt> no it is not, it has not been, and it will not be
2152017-01-17T17:22:13  <luke-jr> I already suggested a compromise that it can be removed at the first occasion of it being a burden to someone who wants to change something there.
2162017-01-17T17:22:21  <BlueMatt> we're waayyyyy past that point
2172017-01-17T17:22:41  <sipa> luke-jr: the discussion about it has dragged on for ages, and just the disagreement about it is draining people who want to work on the project
2182017-01-17T17:22:51  <sipa> it is clear that thid code is useless at this point
2192017-01-17T17:22:54  <BlueMatt> the compromise is that if you want to support it, you can add additional rpcs which give you access to the appropriate data so you can call prioritizetransaction (ie the fee-bumping command)
2202017-01-17T17:22:55  <luke-jr> then morcos went and made that inflammatory comment
2212017-01-17T17:23:06  <sipa> we're doing everyone a favor by just getting past it
2222017-01-17T17:23:08  <luke-jr> sipa: so stop creating disagreement for no reason
2232017-01-17T17:23:34  <BlueMatt> the disagreement is that only you disagree, at this point
2242017-01-17T17:23:34  <sipa> luke-jr: i can say the same
2252017-01-17T17:23:41  <BlueMatt> hell, even other miners disagree with you
2262017-01-17T17:23:53  <luke-jr> except I'm not the one who keeps bringing it up and forcing the issue.
2272017-01-17T17:24:02  <sipa> luke-jr: you're turning a piece of outdated logic from a mouse into an elephant, and i'm tired of having this discussion
2282017-01-17T17:24:04  <BlueMatt> because everyone wants to just remove the damn code already
2292017-01-17T17:24:05  <luke-jr> BlueMatt: some miners perhaps. others don't.
2302017-01-17T17:24:19  <BlueMatt> oh? who other than you?
2312017-01-17T17:24:20  <luke-jr> nobody is forcing the ones who don't want it to use it.
2322017-01-17T17:24:23  <BlueMatt> (and dont say wizkid)
2332017-01-17T17:24:32  <luke-jr> wizkid057 doesn't count now?
2342017-01-17T17:24:39  <BlueMatt> the amount of work that has gone into maintaining this feature just for eligius is insane
2352017-01-17T17:24:46  <BlueMatt> we maintain no other feature for one person/group
2362017-01-17T17:24:52  *** BashCo has joined #bitcoin-core-dev
2372017-01-17T17:24:59  <BlueMatt> if that were our policy we'd still have tonal numbering in bitcoin core
2382017-01-17T17:25:15  <luke-jr> "still" implying we ever did
2392017-01-17T17:25:22  <BlueMatt> well, ok, we'd have
2402017-01-17T17:25:24  <BlueMatt> s/still//
2412017-01-17T17:25:26  <BlueMatt> same thin
2422017-01-17T17:25:27  <BlueMatt> g
2432017-01-17T17:28:56  <luke-jr> not to mention there's no evidence it's only used by one person.
2442017-01-17T17:31:42  <sipa> there is evidence that only a small percentage of the hash rate uses it (look at the feerate of blocks in transactions... it's clearly feerate sorted, except for a small percentage of txn at the beginning of a very occasional block)
2452017-01-17T17:32:06  <sipa> furthermore, it is completely useless as a spam prevention mechanism without wallets targetting it
2462017-01-17T17:32:23  *** laurentmt has quit IRC
2472017-01-17T17:32:42  <sipa> i'm done arguing about this
2482017-01-17T17:32:59  <sipa> we'll remove priority mining in 0.15, as far as i'm concerned
2492017-01-17T17:33:20  <luke-jr> that's simply not true at all. priority works without any targetting.
2502017-01-17T17:34:02  <sipa> come on
2512017-01-17T17:35:22  <sipa> a perfectly legitimate wallet's transaction would be treated completely arbitrarily under it
2522017-01-17T17:35:41  <sipa> all it would accomplish is favor a few people with very old coi s
2532017-01-17T17:36:01  <sipa> until they start spending recent change
2542017-01-17T17:38:04  <luke-jr> *any* age/value weighs more than unconfirmed TXOs.
2552017-01-17T17:38:22  <sipa> if you choose to keep arguing this, i will choose to ignore you
2562017-01-17T17:41:12  <luke-jr> fine, remove priority. be a jerk and do it before there's even a slightly useful purpose in doing so. but then it's time to stop pretending Core is a politically-neutral reference implementation, since it's de facto being used as a vehicle to force network policy. all you're doing is proving the haters right.
2572017-01-17T17:46:30  *** Guyver2 has joined #bitcoin-core-dev
2582017-01-17T17:47:49  *** Victorsueca has joined #bitcoin-core-dev
2592017-01-17T17:49:58  *** laurentmt has joined #bitcoin-core-dev
2602017-01-17T17:50:10  *** Victor_sueca has quit IRC
2612017-01-17T18:00:35  *** chjj has quit IRC
2622017-01-17T18:01:34  <BlueMatt> luke-jr: if there were no way to implement it in the supported apis, maybe you'd have a point...but right now it is /trivial/ to implement priority exactly how you like it in a little python script which loops over the mempool and does manual prioritization
2632017-01-17T18:02:05  <BlueMatt> luke-jr: if you can honestly tell me you've implemented that, and it wasnt performant enough or there was some other issue preventing it from being practical, maybe I'd have some sympathy, but until then....
2642017-01-17T18:08:14  <luke-jr> BlueMatt: so just busyloop over RPC mempool listing, maintain an entire copy of the mempool state outside bitcoind, and act as a proxy for explicit miner prioritisation? I can't even imagine a way to do this effectively; it's certainly a heck of a lot more work, less reliable, and far less efficient.
2652017-01-17T18:08:56  *** Victor_sueca has joined #bitcoin-core-dev
2662017-01-17T18:08:59  <BlueMatt> less efficient? yes, lot more work? I dont believe so...go implement it and prove me wrong
2672017-01-17T18:09:08  <BlueMatt> i dont think it needs state, just iterate over and clear prioritization as you go
2682017-01-17T18:09:10  *** Victorsueca has quit IRC
2692017-01-17T18:10:13  <BlueMatt> high_prio = stack(); for transaction in getrawmempool(): if calculate_priority(tx) is highest in stack: place in stack; for transaction not in stack but previously prioritzed: deprioritize; for transaction in stack: prioritize appropriately
2702017-01-17T18:10:21  <BlueMatt> if thats more than 100 lines of python I'd be surprised
2712017-01-17T18:10:54  <BlueMatt> and if its too ineffecient for you, please add an rpc which provides all the info needed for calculate_priority so that its super effecient
2722017-01-17T18:11:11  <BlueMatt> I'd be more than happy to review that
2732017-01-17T18:13:28  <luke-jr> to have such an RPC means having about half of the priority code stay in Core
2742017-01-17T18:13:40  <luke-jr> because it comes from the input txouts
2752017-01-17T18:13:48  <BlueMatt> dont think so? such an rpc could easily only provide things like foreach input: provide txout info
2762017-01-17T18:14:02  *** Victorsueca has joined #bitcoin-core-dev
2772017-01-17T18:14:03  <BlueMatt> and that is way more general than priority
2782017-01-17T18:14:12  <BlueMatt> it allows folks to implement other crazy rules
2792017-01-17T18:14:17  <BlueMatt> like "spends from address x"
2802017-01-17T18:14:29  <BlueMatt> (well, admittedly you can kinda already do that by seeing the pubkey)
2812017-01-17T18:14:32  <luke-jr> but fetching txouts is where the inefficiency came from?
2822017-01-17T18:14:52  <BlueMatt> fetching txouts is the tightest loop in the above pseudocode
2832017-01-17T18:14:54  <BlueMatt> so I'd think so
2842017-01-17T18:15:17  *** laurentmt has quit IRC
2852017-01-17T18:15:49  <BlueMatt> wumpus: yo
2862017-01-17T18:15:51  *** Victorsueca has quit IRC
2872017-01-17T18:15:52  <BlueMatt> you still around?
2882017-01-17T18:16:11  *** laurentmt has joined #bitcoin-core-dev
2892017-01-17T18:16:40  *** Victor_sueca has quit IRC
2902017-01-17T18:16:59  *** Victorsueca has joined #bitcoin-core-dev
2912017-01-17T18:20:16  *** laurentmt has quit IRC
2922017-01-17T18:20:28  *** isle2983 has joined #bitcoin-core-dev
2932017-01-17T18:24:21  *** chjj has joined #bitcoin-core-dev
2942017-01-17T18:27:53  <luke-jr> BlueMatt: let's continue that topic (or not) post-freeze. either way, it will be easier to maintain priority in Knots than external. and in the meantime, if we haven't frozen yet, I should get back to reviewing stuff
2952017-01-17T18:28:16  <BlueMatt> its unclear to me if we've frozen or not :/
2962017-01-17T18:28:28  <luke-jr> well, finishing up review certainly can't hurt either way
2972017-01-17T18:28:38  <BlueMatt> I mean either we froze, or we have 4 outstnding prs that either go in or dont (todayish, preferably)
2982017-01-17T18:28:42  <BlueMatt> yea, fair point
2992017-01-17T18:32:58  *** isle2983 has quit IRC
3002017-01-17T18:40:30  *** laurentmt has joined #bitcoin-core-dev
3012017-01-17T18:40:43  *** laurentmt has quit IRC
3022017-01-17T18:47:30  *** isle2983 has joined #bitcoin-core-dev
3032017-01-17T18:51:24  *** chjj has quit IRC
3042017-01-17T18:55:07  *** morph has left #bitcoin-core-dev
3052017-01-17T18:55:29  <bitcoin-git> [bitcoin] jnewbery opened pull request #9569: Setting -blocksonly sets -maxmempool to zero. (master...blocksonlynomempoolsharing) https://github.com/bitcoin/bitcoin/pull/9569
3062017-01-17T18:59:15  *** isle2983 has quit IRC
3072017-01-17T19:09:09  *** isle2983 has joined #bitcoin-core-dev
3082017-01-17T19:11:21  *** waxwing has quit IRC
3092017-01-17T19:13:53  *** waxwing has joined #bitcoin-core-dev
3102017-01-17T19:27:37  *** isle2983 has quit IRC
3112017-01-17T19:30:13  *** wasi has quit IRC
3122017-01-17T19:30:36  *** wasi has joined #bitcoin-core-dev
3132017-01-17T19:31:28  *** isle2983 has joined #bitcoin-core-dev
3142017-01-17T19:51:26  *** chjj has joined #bitcoin-core-dev
3152017-01-17T19:52:16  *** paveljanik has joined #bitcoin-core-dev
3162017-01-17T19:52:17  *** paveljanik has joined #bitcoin-core-dev
3172017-01-17T19:59:58  *** Chris_Stewart_5 has quit IRC
3182017-01-17T20:15:48  *** Chris_Stewart_5 has joined #bitcoin-core-dev
3192017-01-17T20:26:05  *** celsosouza has joined #bitcoin-core-dev
3202017-01-17T20:36:41  *** moli_ has joined #bitcoin-core-dev
3212017-01-17T20:37:04  <wumpus> BlueMatt: no, we've not frozen yet, that's when we split off the 0.14 branch
3222017-01-17T20:38:10  *** Sosumi has quit IRC
3232017-01-17T20:39:10  <BlueMatt> wumpus: oh, I'm confused now...the release schedule says feature freeze today, branch in feb
3242017-01-17T20:40:53  <wumpus> BlueMatt: I'm confused apaprently, yes branch is before first -rc, it has been the other way around but it means a lot of more backporting so this is okay
3252017-01-17T20:44:06  <wumpus> the only other freeze there is today is "translation string freeze", but that makes little sense without feature freeze
3262017-01-17T20:44:47  <wumpus> let's move the feature freeze (and translation freeze) to thursday
3272017-01-17T20:46:24  <BlueMatt> my pending list is #8456, #9294 # 9499 and #9535...of those I believe only 9499 has string changes
3282017-01-17T20:46:30  <gribble> https://github.com/bitcoin/bitcoin/issues/8456 | [RPC] Simplified bumpfee command. by mrbandrews · Pull Request #8456 · bitcoin/bitcoin · GitHub
3292017-01-17T20:46:32  <gribble> https://github.com/bitcoin/bitcoin/issues/9294 | Use internal HD chain for change outputs (hd split) by jonasschnelli · Pull Request #9294 · bitcoin/bitcoin · GitHub
3302017-01-17T20:46:34  <gribble> https://github.com/bitcoin/bitcoin/issues/9535 | Split CNode::cs_vSend: message processing and message sending by TheBlueMatt · Pull Request #9535 · bitcoin/bitcoin · GitHub
3312017-01-17T20:47:06  <BlueMatt> I dont think anything else is reasonably gonna make a feature freeze
3322017-01-17T20:47:17  <BlueMatt> so we could also just call translation string freeze when #9499 is merged
3332017-01-17T20:47:20  <gribble> https://github.com/bitcoin/bitcoin/issues/9499 | Use recent-rejects, orphans, and recently-replaced txn for compact-block-reconstruction by TheBlueMatt · Pull Request #9499 · bitcoin/bitcoin · GitHub
3342017-01-17T20:47:36  <BlueMatt> (rpc help text is not translated, correct?)
3352017-01-17T20:47:39  <wumpus> sounds good to me
3362017-01-17T20:47:43  <wumpus> no, rpc help is not translated
3372017-01-17T20:47:48  <BlueMatt> ok, good
3382017-01-17T20:48:44  <wumpus> 9535 isn't even tagged for 0.14.0?!
3392017-01-17T20:49:00  <BlueMatt> no, but is too trivial and big win for me to not include in my list :p
3402017-01-17T20:49:29  <BlueMatt> but if it slips thats ok
3412017-01-17T20:49:31  <wumpus> I think we should focus on the current list, there's still 6 things on there
3422017-01-17T20:49:38  <BlueMatt> as long as 9499 doesnt
3432017-01-17T20:49:43  <wumpus> https://github.com/bitcoin/bitcoin/pulls?q=is%3Aopen+is%3Apr+milestone%3A0.14.0
3442017-01-17T20:49:45  <BlueMatt> half of the current list is bugfixes
3452017-01-17T20:49:54  <BlueMatt> of which there are several more coming (see issues tagged for 14)
3462017-01-17T20:50:15  * BlueMatt goes back to fixing #9148
3472017-01-17T20:50:16  <gribble> https://github.com/bitcoin/bitcoin/issues/9148 | Wallet RPCs can return stale info due to ProcessNewBlock Race · Issue #9148 · bitcoin/bitcoin · GitHub
3482017-01-17T20:50:35  <BlueMatt> which is nontrivial :(
3492017-01-17T20:50:35  <wumpus> bumpfee, internal hd chain, fundrawtransaction, improve progress display are all features I'd say
3502017-01-17T20:50:48  <BlueMatt> oh fuck, how did i miss the display one
3512017-01-17T20:51:01  <BlueMatt> no, the fundraw one is def a bugfix
3522017-01-17T20:51:05  <wumpus> exclude RBF replacement and recent-rejects seem bugfixes
3532017-01-17T20:51:16  <BlueMatt> recent-rejects is more a feature
3542017-01-17T20:51:20  <wumpus> ok
3552017-01-17T20:51:52  <BlueMatt> without the fundarw one it re-uses change addresses
3562017-01-17T20:52:00  <BlueMatt> which is very not so good
3572017-01-17T20:52:20  <BlueMatt> fuck, jonasschnelli #9377 needs rebase
3582017-01-17T20:52:21  <wumpus> I slotted it as feature as it adds to the RPC API
3592017-01-17T20:52:21  <gribble> https://github.com/bitcoin/bitcoin/issues/9377 | fundrawtransaction: Keep change-output keys by default, make it optional by jonasschnelli · Pull Request #9377 · bitcoin/bitcoin · GitHub
3602017-01-17T20:52:41  <BlueMatt> oh it does add
3612017-01-17T20:52:41  <wumpus> anyhow it's important to get in that's for sure
3622017-01-17T20:52:49  <BlueMatt> hum, we could probably implement it more cleanly
3632017-01-17T20:52:58  <BlueMatt> well, cleanly meaning less diff
3642017-01-17T20:53:01  <BlueMatt> but, yea, needs to happen for 14
3652017-01-17T20:53:07  <wumpus> is there time for that and still review the stuff before 0.14?
3662017-01-17T20:53:31  <BlueMatt> my concern for 14 isnt the open stuff, its the bugs that need fixing that arent even pr'ed yet
3672017-01-17T20:53:43  <BlueMatt> eg #9148 is a serious issue and the fix is not tiny
3682017-01-17T20:53:44  <gribble> https://github.com/bitcoin/bitcoin/issues/9148 | Wallet RPCs can return stale info due to ProcessNewBlock Race · Issue #9148 · bitcoin/bitcoin · GitHub
3692017-01-17T20:53:48  <wumpus> the diff for #9377 is not that large
3702017-01-17T20:53:50  <gribble> https://github.com/bitcoin/bitcoin/issues/9377 | fundrawtransaction: Keep change-output keys by default, make it optional by jonasschnelli · Pull Request #9377 · bitcoin/bitcoin · GitHub
3712017-01-17T20:54:01  <BlueMatt> yea, I'm not too worried about that one, will review soon
3722017-01-17T20:54:47  <BlueMatt> I also plan to do a helgrind run again this or next week, whenever we decide which net pulls will make it, which is gonna result in a nontrivial number of "Convert X to std::atomic" commits
3732017-01-17T20:55:56  <BlueMatt> the walletnotify one (#9479 / $9371) is also gonna take some work
3742017-01-17T20:55:57  <gribble> https://github.com/bitcoin/bitcoin/issues/9479 | An error has occurred and has been logged. Please contact this bot's administrator for more information.
3752017-01-17T20:56:09  *** celsosouza has quit IRC
3762017-01-17T20:56:21  <BlueMatt> but i havent dug into that one as of yet
3772017-01-17T20:56:27  <BlueMatt> morcos and y'all seem on it
3782017-01-17T20:57:59  <BlueMatt> ok, so string freeze is whenever #9499 and #9461 make it, otherwise when feature freeze happens
3792017-01-17T20:58:02  <gribble> https://github.com/bitcoin/bitcoin/issues/9499 | Use recent-rejects, orphans, and recently-replaced txn for compact-block-reconstruction by TheBlueMatt · Pull Request #9499 · bitcoin/bitcoin · GitHub
3802017-01-17T20:58:04  <gribble> https://github.com/bitcoin/bitcoin/issues/9461 | [Qt] Improve progress display during headers-sync and peer-finding by jonasschnelli · Pull Request #9461 · bitcoin/bitcoin · GitHub
3812017-01-17T20:58:09  *** handlex has joined #bitcoin-core-dev
3822017-01-17T20:58:15  <BlueMatt> jonasschnelli: can you fix the outstanding comments on #9461?
3832017-01-17T20:58:17  <gribble> https://github.com/bitcoin/bitcoin/issues/9461 | [Qt] Improve progress display during headers-sync and peer-finding by jonasschnelli · Pull Request #9461 · bitcoin/bitcoin · GitHub
3842017-01-17T21:05:38  *** cryptapus has quit IRC
3852017-01-17T21:10:35  *** waxwing has quit IRC
3862017-01-17T21:25:25  <BlueMatt> 'tf
3872017-01-17T21:25:47  <BlueMatt> so I'm pretty sure, right now, if you receive a coinbase payout in a block, turn off your node, and then restart without checkblocks, your payout will not display
3882017-01-17T21:26:23  <BlueMatt> because no NotifyTransactionChanged will ever fire to the gui
3892017-01-17T21:27:16  <BlueMatt> i mean super edge-case-y but still
3902017-01-17T21:32:14  <sipa> is that a recent regression?
3912017-01-17T21:33:29  <BlueMatt> looks super old
3922017-01-17T21:33:33  <BlueMatt> but didnt check
3932017-01-17T21:33:39  <BlueMatt> restarting gui will fix it, though, afaict
3942017-01-17T22:12:53  *** handlex has quit IRC
3952017-01-17T22:13:46  *** handlex has joined #bitcoin-core-dev
3962017-01-17T22:29:38  *** handlex has quit IRC
3972017-01-17T22:39:05  *** jannes has quit IRC
3982017-01-17T23:08:56  *** aalex has quit IRC
3992017-01-17T23:18:28  *** handlex has joined #bitcoin-core-dev
4002017-01-17T23:51:34  *** Guyver2 has quit IRC
4012017-01-17T23:53:52  <bitcoin-git> [bitcoin] TheBlueMatt opened pull request #9570: Block Wallet RPCs until wallet is synced to our current chain (master...2017-01-fix-wallet-rpc-stale) https://github.com/bitcoin/bitcoin/pull/9570
4022017-01-17T23:56:53  <gmaxwell> I thought the fundraw change change made it in already. darn.
4032017-01-17T23:59:08  *** MarcoFalke has joined #bitcoin-core-dev