1 2017-03-24 00:01:07	0|jtimon|so in case they don't do it with segwit inactive is a bug, otherwise is expected behaviour, right?
  2 2017-03-24 00:04:32	0|jtimon|english speakers: sorry for the double negations, I didn't lacked no ineducation
  3 2017-03-24 00:10:31	0|gmaxwell|jtimon: segwit is active on testnet.
  4 2017-03-24 00:11:11	0|gmaxwell|It's expected behavior on testnet that nodes will not fetch blocks from non-segwit peers, because non-segwit peers cannot provide witnesses, which are required, since segwit is active.
  5 2017-03-24 00:11:53	0|jtimon|oh, right, the assumptions to reproduce the error aren't even possible on testnet3 right now, thanks for clarifying
  6 2017-03-24 00:12:04	0|sipa|jtimon: ?
  7 2017-03-24 00:12:17	0|sipa|what is 'the error' you're referring to?
  8 2017-03-24 00:14:02	0|jtimon|the potential error was segwit nodes not downloading blocks from non-segwit nodes when segwit is not active, which cannot be reproduced in testnet3 as pointed out by greg
  9 2017-03-24 00:14:21	0|sipa|nobody is talking about the case where segwit is not active
 10 2017-03-24 00:14:33	0|jtimon|sorry then
 11 2017-03-24 00:22:20	0|jtimon|I can't believe I didn't mentioned #9279, next meeting
 12 2017-03-24 00:22:22	0|gribble|https://github.com/bitcoin/bitcoin/issues/9279 | Consensus: Move CFeeRate out of libconsensus by jtimon · Pull Request #9279 · bitcoin/bitcoin · GitHub
 13 2017-03-24 01:00:03	0|gmaxwell|luke-jr: can you explain to me how your size proof deals with blocksizes that aren't a power of two. I'm too intellectually lazy to read that out of your document, and its not obvious to me how you'd deal with it except something like "don't count the minimum size for segments of the tree which are elided so that you can't look for duplication.
 14 2017-03-24 01:58:53	0|phantomcircuit|gmaxwell, we're only using openssl for secp256k1 tests now right?
 15 2017-03-24 01:59:04	0|sipa|and as RNG
 16 2017-03-24 01:59:09	0|phantomcircuit|oh right
 17 2017-03-24 01:59:16	0|gmaxwell|and for payment protocol in QT
 18 2017-03-24 01:59:16	0|sipa|and in payment protocol processing
 19 2017-03-24 01:59:22	0|gribble|Error: "hi5" is not a valid command.
 20 2017-03-24 01:59:22	0|sipa|!hi5
 21 2017-03-24 01:59:26	0|phantomcircuit|yeah so openssl appears to be trying to change their license in a way that appears uh
 22 2017-03-24 01:59:28	0|phantomcircuit|questionable
 23 2017-03-24 01:59:38	0|phantomcircuit|iono just mentioning it
 24 2017-03-24 02:00:34	0|achow101|phantomcircuit: this https://www.openssl.org/blog/blog/2017/03/22/license/ ?
 25 2017-03-24 02:01:27	0|achow101|they're moving to apache 2.0
 26 2017-03-24 02:03:08	0|phantomcircuit|achow101, > If we do not hear from you, we will assume that you have no objection.
 27 2017-03-24 02:03:10	0|phantomcircuit|k
 28 2017-03-24 02:04:49	0|achow101|where do you see that?
 29 2017-03-24 02:06:02	0|gmaxwell|yea, thats ... uh.
 30 2017-03-24 02:13:28	0|phantomcircuit|achow101, it's in the email they sent to all contributors
 31 2017-03-24 02:13:36	0|luke-jr|gmaxwell: the minimum 60 bytes is only assumed up to the last full tx proof, or 1 greater than the half-way point
 32 2017-03-24 02:14:55	0|gmaxwell|luke-jr: couldn't I just create a block that had 2^n+1 transactions  for some n and then provide a full proof for the 'last' transaction (really one long past the end).
 33 2017-03-24 02:15:35	0|luke-jr|gmaxwell: that's the part I haven't solved yet. I *think* it's sufficient to simply forbid any duplicate hashes building to the root?
 34 2017-03-24 02:16:16	0|sipa|yup
 35 2017-03-24 02:16:27	0|sipa|that's exactly what you should do
 36 2017-03-24 02:16:41	0|gmaxwell|luke-jr: yes, but you don't see the whole tree. I think it's sufficient to not count minimum size for txn where you possibly couldn't see duplication.
 37 2017-03-24 02:17:12	0|gmaxwell|e.g. say you should a full proof for the coinbase, and nothing else, but perhaps the last 1/4th of the block was a duplicated branch.
 38 2017-03-24 02:17:33	0|luke-jr|that's why only 1/2+1 gets the min size counted
 39 2017-03-24 02:17:39	0|gmaxwell|if you only show the hashes along the path to the coinbase you could only count half the number of minimum transactions.
 40 2017-03-24 02:17:57	0|gmaxwell|okay. sounds good.
 41 2017-03-24 02:18:24	0|luke-jr|if you prove the full tx at/near the end, though, you'd encounter a duplicate hash somewhere building upward
 42 2017-03-24 02:19:05	0|gmaxwell|yes, I think thats right, okay, so you can count all minimum up to the last shown transaction, and half that beyond that.
 43 2017-03-24 02:19:19	0|gmaxwell|that does make my proposed optimization process suboptimal.
 44 2017-03-24 02:20:12	0|luke-jr|can you count half beyond it? I was assuming min(half+1, lastFullProven)
 45 2017-03-24 02:20:17	0|gmaxwell|(since there is an advantage in making one of the full txn shown be late in the block.
 46 2017-03-24 02:21:02	0|luke-jr|it's probably better than what I have now
 47 2017-03-24 02:21:11	0|luke-jr|currently I'm just adding them largest first
 48 2017-03-24 04:35:38	0|bitcoin-git|[13bitcoin] 15NicolasDorier opened pull request #10068: [WIP] FundRawTransaction accepts preset non-wallet inputs (06master...06nonwalletinputs) 02https://github.com/bitcoin/bitcoin/pull/10068
 49 2017-03-24 06:57:00	0|NicolasDorier|is there a C++ guru in the room ?? I am having this mem fault:
 50 2017-03-24 06:57:04	0|NicolasDorier|https://www.irccloud.com/pastebin/1K7ZnV86/
 51 2017-03-24 06:57:18	0|NicolasDorier|after doing what I thought pretty harmless
 52 2017-03-24 06:57:30	0|NicolasDorier|txout = anotherTxOut
 53 2017-03-24 06:57:58	0|NicolasDorier|it seems that it is not possible because CScript operator= is exploding
 54 2017-03-24 06:58:22	0|NicolasDorier|which kind of amaze me, as I don't see how such a thing is possible
 55 2017-03-24 07:04:50	0|NicolasDorier|https://www.irccloud.com/pastebin/uJAM5yEn/
 56 2017-03-24 07:09:39	0|NicolasDorier|realloc invalid pointer
 57 2017-03-24 07:09:47	0|dcousens|NicolasDorier: aye, trying to figure it out myself
 58 2017-03-24 07:09:48	0|NicolasDorier|the pointer not being null
 59 2017-03-24 07:12:09	0|NicolasDorier|https://www.irccloud.com/pastebin/RG0eniJZ/
 60 2017-03-24 07:13:00	0|wumpus|BlueMatt: yes I'll create a project (if no one has done so yet)
 61 2017-03-24 07:14:46	0|luke-jr|NicolasDorier: run it in valgrind if you can; random memory corruption can manifest in weird ways
 62 2017-03-24 07:14:56	0|luke-jr|NicolasDorier: pastebin the patch? :p
 63 2017-03-24 07:17:50	0|NicolasDorier|luke-jr: My code https://github.com/bitcoin/bitcoin/pull/10068 the innocent looking faulty line is https://github.com/bitcoin/bitcoin/pull/10068/files#diff-b2bb174788c7409b671c46ccc86034bdL2118 that you can hit by running sendtoaddress
 64 2017-03-24 07:18:13	0|NicolasDorier|I don't know valgrind, better than gdb ?
 65 2017-03-24 07:20:01	0|NicolasDorier|dcousens: might be some unrelated problem of memory corruption ? I will try to make an easy repro
 66 2017-03-24 07:20:16	0|luke-jr|NicolasDorier: valgrind is for memory stuff
 67 2017-03-24 07:20:19	0|NicolasDorier|it fails deterministicallly here though
 68 2017-03-24 07:20:21	0|NicolasDorier|ha
 69 2017-03-24 07:20:27	0|NicolasDorier|ok I'll try that
 70 2017-03-24 07:31:43	0|NicolasDorier|dcousens: seems the problem was my code.
 71 2017-03-24 07:31:48	0|NicolasDorier|https://www.irccloud.com/pastebin/tmHUV2sU/
 72 2017-03-24 07:31:55	0|NicolasDorier|I was in the thought that
 73 2017-03-24 07:32:06	0|NicolasDorier|the two fields
 74 2017-03-24 07:32:15	0|NicolasDorier|would be initialized by their default ctor
 75 2017-03-24 07:32:23	0|NicolasDorier|and thus set to Null
 76 2017-03-24 07:32:51	0|NicolasDorier|I mean SetNull would be called on them
 77 2017-03-24 07:33:52	0|NicolasDorier|luke-jr: awesome tool thanks.
 78 2017-03-24 07:49:45	0|NicolasDorier|nop atcually was not the problem
 79 2017-03-24 07:53:12	0|wumpus|created the project. Any proposed pulls from last meeting (or the one before that) that I missed? https://github.com/bitcoin/bitcoin/projects/8
 80 2017-03-24 08:07:20	0|NicolasDorier|wumpus: proposed pulls for what ?
 81 2017-03-24 08:07:45	0|wumpus|NicolasDorier: for priority review (some people were asking for this at the meeting)
 82 2017-03-24 08:08:21	0|NicolasDorier|ah ok thought it was about proposals for 0.15
 83 2017-03-24 08:08:48	0|wumpus|well, in a way it is, but 0.15 is still too far away to have a review crunch now
 84 2017-03-24 08:09:16	0|wumpus|this is not about last minute merges to any release but about what is blocking people now
 85 2017-03-24 08:09:45	0|NicolasDorier|wumpus: the merge of the leveldb fix to bitcoin
 86 2017-03-24 08:10:04	0|NicolasDorier|well no actually it is not like it blocks me
 87 2017-03-24 08:43:51	0|NicolasDorier|AH I found out my bug. Very interesting. My class was named CCoin, but there is another CCoin in rest.cpp
 88 2017-03-24 08:43:53	0|NicolasDorier|strangely
 89 2017-03-24 08:44:07	0|NicolasDorier|I had no compile error problem
 90 2017-03-24 08:44:26	0|NicolasDorier|even if both CCoin had completely different methods and fields
 91 2017-03-24 08:44:31	0|NicolasDorier|but on runtime
 92 2017-03-24 08:44:46	0|NicolasDorier|the other CCoin was linked oO
 93 2017-03-24 08:45:13	0|NicolasDorier|so basically what I compiled was not what was linked or some strange stuff like that
 94 2017-03-24 08:45:44	0|NicolasDorier|changing the name of the class completely fixed the problem
 95 2017-03-24 08:49:29	0|luke-jr|O.o
 96 2017-03-24 09:27:32	0|dcousens|NicolasDorier: o.O
 97 2017-03-24 09:28:11	0|NicolasDorier|yeah... what tip me off is when I was using gdb, but the debugger was going into the CCoin of rest.cpp
 98 2017-03-24 09:28:38	0|NicolasDorier|no idea how it could compile since after I was calling methhods only from my new CCoin class
 99 2017-03-24 09:29:11	0|dcousens|NicolasDorier: CCoin::operator= (this=0x7fffdb7fc9f0) at ./wallet/wallet.h:459 -  is a blank line for me, guessing you had modifications there for sure
100 2017-03-24 09:29:31	0|dcousens|esp. since CCoin doesn't even have a operator= in that file
101 2017-03-24 09:29:32	0|NicolasDorier|operator= was not defined
102 2017-03-24 09:29:42	0|NicolasDorier|it was the default operator= which should just copy
103 2017-03-24 09:29:55	0|NicolasDorier|in C++ it exists by default
104 2017-03-24 09:30:19	0|sipa|NicolasDorier: if you'd have had the two CCoin class definitions included within one module, compilation would fail
105 2017-03-24 09:30:44	0|NicolasDorier|rest.h was not included into wallet.cpp where I was using my CCoin
106 2017-03-24 09:30:55	0|NicolasDorier|so there wwas no conflict as far as wallet.cpp was concerned
107 2017-03-24 09:32:14	0|sipa|and neither of these has virtual methods?
108 2017-03-24 09:32:48	0|sipa|i'm confused actually... if it compiled, i would expect it to work
109 2017-03-24 09:32:57	0|NicolasDorier|nop
110 2017-03-24 09:33:11	0|dcousens|NicolasDorier: are you certain that was the issue? not just some compilational coincidence?
111 2017-03-24 09:33:14	0|NicolasDorier|yes... well I was bit surprised as well
112 2017-03-24 09:33:19	0|sipa|the linker would have failed if there was a duplicate definition for some method that wasn't inlined
113 2017-03-24 09:33:35	0|sipa|and if they were inlined, the one you expect would be inlined
114 2017-03-24 09:33:42	0|NicolasDorier|dcousens: ah maybe a .o outdated or something like that ?
115 2017-03-24 09:33:52	0|NicolasDorier|I indeed did not made a make clean
116 2017-03-24 09:34:00	0|dcousens|NicolasDorier: indeed, it was going to be my first suggestion `make clean` haha
117 2017-03-24 09:34:26	0|NicolasDorier|mmh possible. But strange, it is not like I played with ./configure :/
118 2017-03-24 09:34:52	0|NicolasDorier|my methods were inlined
119 2017-03-24 09:38:33	0|luke-jr|sipa: would it? I would think the same signature would overwrite silently (and that this is relied on for templates)?
120 2017-03-24 09:44:06	0|luke-jr|otoh, I think I may have seen errors like that before
121 2017-03-24 10:06:57	0|jouke|~/win 24
122 2017-03-24 10:07:00	0|jouke|srsly
123 2017-03-24 10:07:20	0|jouke|I need to get an other irc client.
124 2017-03-24 10:11:01	0|bitcoin-git|[13bitcoin] 15NicolasDorier opened pull request #10069: [QA] Fix typo in fundrawtransaction test (06master...06patch-3) 02https://github.com/bitcoin/bitcoin/pull/10069
125 2017-03-24 11:47:25	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #10051: adhere to `-whitelist` for outbound connection (06master...06whitelist-outbound) 02https://github.com/bitcoin/bitcoin/pull/10051
126 2017-03-24 11:49:13	0|bitcoin-git|[13bitcoin] 15laanwj reopened pull request #10051: adhere to `-whitelist` for outbound connection (06master...06whitelist-outbound) 02https://github.com/bitcoin/bitcoin/pull/10051
127 2017-03-24 11:49:44	0|dcousens|NicolasDorier: beat me to saying it
128 2017-03-24 11:49:48	0|dcousens|`rawtx` and `rawTx`
129 2017-03-24 11:49:50	0|dcousens|lol
130 2017-03-24 11:50:10	0|NicolasDorier|took me an hour to understand wtf was going on
131 2017-03-24 11:50:24	0|dcousens|can they not be broken out into different scope?
132 2017-03-24 11:50:28	0|dcousens|i mean, different PR, but still
133 2017-03-24 11:51:16	0|NicolasDorier|what do you mean ?
134 2017-03-24 11:52:00	0|dcousens|NicolasDorier: save re-writing the whole thing, I wonder if you could do the equivalent of introducing a block scope
135 2017-03-24 11:52:45	0|NicolasDorier|rewriting the whole thing ?
136 2017-03-24 11:52:55	0|dcousens|NicolasDorier: dw, I just meant to avoid the error in future
137 2017-03-24 11:52:56	0|NicolasDorier|I don't follow you
138 2017-03-24 11:53:24	0|NicolasDorier|ah I did not knew block scope existed in python actually :p
139 2017-03-24 11:53:28	0|dcousens|`rawtx` ideally would have thrown "undefined" error if there was an actual typo
140 2017-03-24 11:53:43	0|dcousens|NicolasDorier: iirc they are function scoped anyway
141 2017-03-24 11:54:06	0|wumpus|block scope doesn't exist inpython just function scope
142 2017-03-24 11:54:41	0|NicolasDorier|ha it is possible if broken into functions yes. But I think just better to use rawtx or rawTx everywhere consistently
143 2017-03-24 11:54:42	0|wumpus|(although there are "with ..." constructs that can be used to do a kind of block scope, this is not what they're intended for :-)
144 2017-03-24 11:54:51	0|wumpus|yes just use a consistent name
145 2017-03-24 11:55:02	0|NicolasDorier|might do that in a separate PR
146 2017-03-24 11:55:09	0|dcousens|wumpus: indeed that was my second thought
147 2017-03-24 12:12:13	0|wumpus|I'd prefer to do it in the same PR, just fix the underlying issue that caused this typo in the same go
148 2017-03-24 12:12:29	0|wumpus|I much prefer that to just fixing a symptom :)
149 2017-03-24 12:40:47	0|jonasschnelli|NicolasDorier: can you write a description/motivation for #10068?
150 2017-03-24 12:40:49	0|gribble|https://github.com/bitcoin/bitcoin/issues/10068 | [Wallet] FundRawTransaction accepts preset non-wallet inputs by NicolasDorier · Pull Request #10068 · bitcoin/bitcoin · GitHub
151 2017-03-24 12:41:04	0|NicolasDorier|Yes I am doing right now
152 2017-03-24 12:44:54	0|bitcoin-git|[13bitcoin] 15laanwj closed pull request #9983: tests: Convert selected tests to using named RPC arguments (06master...062017_03_rpc_tests_named_arguments) 02https://github.com/bitcoin/bitcoin/pull/9983
153 2017-03-24 12:52:02	0|NicolasDorier|jonasschnelli: Just did
154 2017-03-24 12:53:39	0|jonasschnelli|NicolasDorier: I haven't really understand the use case so far.. but what's the reason for using fundraw if it's not your inputs? Why not directly add the inputs in createraw?
155 2017-03-24 12:54:04	0|jonasschnelli|IMO fundraw can topup a tx with some pre-filled inputs
156 2017-03-24 12:54:22	0|NicolasDorier|jonasschnelli: right now fundraw can't topup a tx with pre filled inputs
157 2017-03-24 12:54:36	0|NicolasDorier|if the pre filled inputs are not belonging to the wallet
158 2017-03-24 12:54:39	0|jonasschnelli|NicolasDorier: oh... I think my initial version did support that...
159 2017-03-24 12:54:52	0|jonasschnelli|Ah... that's a point
160 2017-03-24 12:55:13	0|NicolasDorier|jonasschnelli: https://github.com/bitcoin/bitcoin/pull/10068/files#diff-b2bb174788c7409b671c46ccc86034bdL2213
161 2017-03-24 12:55:18	0|NicolasDorier|it was in a TODO
162 2017-03-24 12:55:53	0|jonasschnelli|Heh.. at least there was once a plan to do that... okay. I got your point. Thanks for explaining.
163 2017-03-24 12:59:54	0|NicolasDorier|I added "pre filled input" in my description
164 2017-03-24 12:59:58	0|NicolasDorier|it was indeed unclear
165 2017-03-24 13:00:22	0|bitcoin-git|13bitcoin/06master 14232b666 15John Newbery: Allow test cases to be skipped...
166 2017-03-24 13:00:22	0|bitcoin-git|[13bitcoin] 15MarcoFalke pushed 3 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/a230b0588788...ca209230c8e7
167 2017-03-24 13:00:23	0|bitcoin-git|13bitcoin/06master 140c1ade6 15John Newbery: Skip rpcbind_test if OS/network requirements are not met.
168 2017-03-24 13:00:23	0|bitcoin-git|13bitcoin/06master 14ca20923 15MarcoFalke: Merge #10053: [test] Allow functional test cases to be skipped...
169 2017-03-24 13:00:42	0|bitcoin-git|[13bitcoin] 15MarcoFalke closed pull request #10053: [test] Allow functional test cases to be skipped (06master...06skiptests) 02https://github.com/bitcoin/bitcoin/pull/10053
170 2017-03-24 13:05:30	0|bitcoin-git|13bitcoin/06master 1488e3aa0 15John Newbery: Run extended tests once daily
171 2017-03-24 13:05:30	0|bitcoin-git|[13bitcoin] 15MarcoFalke pushed 2 new commits to 06master: 02https://github.com/bitcoin/bitcoin/compare/ca209230c8e7...a0b1e57b20a1
172 2017-03-24 13:05:31	0|bitcoin-git|13bitcoin/06master 14a0b1e57 15MarcoFalke: Merge #10052: [test] Run extended tests once daily in Travis...
173 2017-03-24 13:05:53	0|bitcoin-git|[13bitcoin] 15MarcoFalke closed pull request #10052: [test] Run extended tests once daily in Travis (06master...06traviscronjobs) 02https://github.com/bitcoin/bitcoin/pull/10052
174 2017-03-24 20:29:21	0|bitcoin-git|[13bitcoin] 15jnewbery opened pull request #10072: Remove sources of unreliablility in extended functional tests (06master...06extended_test_unreliablility) 02https://github.com/bitcoin/bitcoin/pull/10072
175 2017-03-24 20:31:25	0|cfields|BlueMatt: ping
176 2017-03-24 20:31:34	0|BlueMatt|yo
177 2017-03-24 20:32:29	0|cfields|BlueMatt: looks like #9605 is hitting the cpu hard
178 2017-03-24 20:32:31	0|gribble|https://github.com/bitcoin/bitcoin/issues/9605 | Use CScheduler for wallet flushing, remove ThreadFlushWalletDB by TheBlueMatt · Pull Request #9605 · bitcoin/bitcoin · GitHub
179 2017-03-24 20:32:36	0|cfields|not sure how i didn't notice pre-merge :(
180 2017-03-24 20:32:44	0|BlueMatt|thats...strange
181 2017-03-24 20:32:44	0|BlueMatt|wtf
182 2017-03-24 20:32:56	0|BlueMatt|grrr, if there's more bugs in cscheduler.....
183 2017-03-24 20:33:02	0|BlueMatt|I assume its in the scheduler thread?
184 2017-03-24 20:33:47	0|cfields|unsure where the overhead is, figured i'd poke you first for ideas
185 2017-03-24 20:34:14	0|cfields|i'm not managing to hit it while it's busy with gdb, so not sure yet
186 2017-03-24 20:34:29	0|BlueMatt|you should be able to see it with top H
187 2017-03-24 20:34:37	0|BlueMatt|(if its scheduler thread)
188 2017-03-24 20:35:38	0|cfields|cool!
189 2017-03-24 20:35:40	0|cfields|yep
190 2017-03-24 20:35:55	0|cfields|(how have i not known this?!)
191 2017-03-24 20:36:11	0|BlueMatt|i dont know
192 2017-03-24 20:38:04	0|sipa|what? top supports threads?
193 2017-03-24 20:38:09	0|BlueMatt|yes!
194 2017-03-24 20:38:10	0|sipa|TIL.
195 2017-03-24 20:38:15	0|BlueMatt|press H while its running to flip
196 2017-03-24 20:38:20	0|sipa|i always used htop for that
197 2017-03-24 22:26:07	0|bitcoin-git|[13bitcoin] 15jnewbery opened pull request #10073: Actually run assumevalid.py (06master...06improveassumevalid) 02https://github.com/bitcoin/bitcoin/pull/10073