1 2017-06-05T00:00:18 *** lightningbot has joined #bitcoin-core-dev 2 2017-06-05T00:00:31 *** davec_ has joined #bitcoin-core-dev 3 2017-06-05T00:00:47 *** d_t has joined #bitcoin-core-dev 4 2017-06-05T00:01:38 *** thermoman_ has joined #bitcoin-core-dev 5 2017-06-05T00:01:49 *** thermoman has quit IRC 6 2017-06-05T00:01:52 *** chjj has joined #bitcoin-core-dev 7 2017-06-05T00:06:19 *** davec has quit IRC 8 2017-06-05T00:31:46 *** Dyaheon has quit IRC 9 2017-06-05T00:32:33 *** Dyaheon has joined #bitcoin-core-dev 10 2017-06-05T00:47:12 *** btcdrak has joined #bitcoin-core-dev 11 2017-06-05T00:49:00 *** Ylbam has quit IRC 12 2017-06-05T01:14:52 *** cryptapus has joined #bitcoin-core-dev 13 2017-06-05T01:14:58 *** cryptapus is now known as cryptapus_afk 14 2017-06-05T01:18:02 *** beatrootfarmer has joined #bitcoin-core-dev 15 2017-06-05T01:22:11 *** goatturneer has quit IRC 16 2017-06-05T01:23:46 <NicolasDorier> spudowiar: Awesome. I think you should share that as well with Ledger (btchip), they would probably review and make a plugin. Also it might move forward the conversation about HW standard 17 2017-06-05T01:36:56 <bitcoin-git> [bitcoin] fanquake closed pull request #10513: Trivial: grammar fix to CONTRIBUTING.md (master...patch-1) https://github.com/bitcoin/bitcoin/pull/10513 18 2017-06-05T02:00:08 *** dermoth has quit IRC 19 2017-06-05T02:01:00 *** dermoth has joined #bitcoin-core-dev 20 2017-06-05T02:18:54 *** Aaronvan_ has quit IRC 21 2017-06-05T02:19:28 *** AaronvanW has joined #bitcoin-core-dev 22 2017-06-05T02:24:15 *** AaronvanW has quit IRC 23 2017-06-05T02:25:38 *** RubenSomsen has joined #bitcoin-core-dev 24 2017-06-05T02:26:35 *** RubenSomsen has quit IRC 25 2017-06-05T02:26:54 *** RubenSomsen has joined #bitcoin-core-dev 26 2017-06-05T02:36:25 *** Dyaheon has quit IRC 27 2017-06-05T02:37:09 *** Dyaheon has joined #bitcoin-core-dev 28 2017-06-05T02:39:57 *** cryptapus_afk has quit IRC 29 2017-06-05T02:44:06 *** nemgun has quit IRC 30 2017-06-05T03:09:50 *** goatturneer has joined #bitcoin-core-dev 31 2017-06-05T03:14:17 *** beatrootfarmer has quit IRC 32 2017-06-05T03:19:00 *** elkalamar has quit IRC 33 2017-06-05T03:22:57 *** beatrootfarmer has joined #bitcoin-core-dev 34 2017-06-05T03:24:24 *** goatturner has joined #bitcoin-core-dev 35 2017-06-05T03:26:59 *** goatturneer has quit IRC 36 2017-06-05T03:28:13 *** beatrootfarmer has quit IRC 37 2017-06-05T04:05:02 *** d9b4bef9 has quit IRC 38 2017-06-05T04:06:08 *** d9b4bef9 has joined #bitcoin-core-dev 39 2017-06-05T04:07:05 *** goatturneer has joined #bitcoin-core-dev 40 2017-06-05T04:10:51 *** goatturner has quit IRC 41 2017-06-05T04:43:31 *** Guest34260 has quit IRC 42 2017-06-05T05:07:28 *** RubenSomsen has quit IRC 43 2017-06-05T05:13:56 *** beatrootfarmer has joined #bitcoin-core-dev 44 2017-06-05T05:17:35 *** goatturneer has quit IRC 45 2017-06-05T05:20:50 *** AaronvanW has joined #bitcoin-core-dev 46 2017-06-05T05:25:10 *** AaronvanW has quit IRC 47 2017-06-05T05:29:11 *** goatturneer has joined #bitcoin-core-dev 48 2017-06-05T05:33:27 *** beatrootfarmer has quit IRC 49 2017-06-05T06:16:42 *** spudowiar has joined #bitcoin-core-dev 50 2017-06-05T06:17:44 <spudowiar> NicolasDorier: Will do. I think I'll make a new RPC method (signhwwtransaction) first (because this isn't entirely compatible with signrawtransaction) and then ask him to review it 51 2017-06-05T06:21:35 *** spudowiar has quit IRC 52 2017-06-05T06:44:53 <bitcoin-git> [bitcoin] laanwj pushed 1 new commit to 0.14: https://github.com/bitcoin/bitcoin/commit/21e1ed48989e1d150e9f85770fd098534e0f81f1 53 2017-06-05T06:44:53 <bitcoin-git> bitcoin/0.14 21e1ed4 Wladimir J. van der Laan: doc: Preliminary release notes 0.14.2 54 2017-06-05T06:50:10 *** lightningbot has joined #bitcoin-core-dev 55 2017-06-05T06:50:26 *** d_t has joined #bitcoin-core-dev 56 2017-06-05T06:50:27 *** wallet42 has joined #bitcoin-core-dev 57 2017-06-05T06:50:46 *** mappum has joined #bitcoin-core-dev 58 2017-06-05T06:58:34 <NicolasDorier> spudowiar: what? signrawtransaction should just work. I mean the wallet already have all the info needed for a hardware wallet to sign 59 2017-06-05T06:58:49 <NicolasDorier> so why a new rpc method needed ? oO 60 2017-06-05T06:59:44 <NicolasDorier> the cool thing if we can reuse signrawtransaction is that moving from software hotwallet to hardware wallet is a configuration change, not a code change for users 61 2017-06-05T07:05:44 <NicolasDorier> Confirmed devs right now are 62 2017-06-05T07:21:30 *** AaronvanW has joined #bitcoin-core-dev 63 2017-06-05T07:24:02 *** d9b4bef9 has quit IRC 64 2017-06-05T07:25:08 *** d9b4bef9 has joined #bitcoin-core-dev 65 2017-06-05T07:26:24 *** AaronvanW has quit IRC 66 2017-06-05T07:32:04 *** Giszmo has joined #bitcoin-core-dev 67 2017-06-05T07:48:09 *** altoz_ has joined #bitcoin-core-dev 68 2017-06-05T07:50:17 *** altoz has quit IRC 69 2017-06-05T07:54:56 *** paveljanik has joined #bitcoin-core-dev 70 2017-06-05T07:54:56 *** paveljanik has joined #bitcoin-core-dev 71 2017-06-05T08:10:05 *** ChillazZ has joined #bitcoin-core-dev 72 2017-06-05T08:11:12 *** AaronvanW has joined #bitcoin-core-dev 73 2017-06-05T08:12:40 *** Ylbam has joined #bitcoin-core-dev 74 2017-06-05T08:26:23 *** jtimon has joined #bitcoin-core-dev 75 2017-06-05T08:31:03 *** timothy has joined #bitcoin-core-dev 76 2017-06-05T08:39:59 *** timothy has quit IRC 77 2017-06-05T08:44:59 *** timothy has joined #bitcoin-core-dev 78 2017-06-05T08:46:10 *** vicenteH has joined #bitcoin-core-dev 79 2017-06-05T08:51:23 *** cryptapus has joined #bitcoin-core-dev 80 2017-06-05T08:51:23 *** cryptapus has joined #bitcoin-core-dev 81 2017-06-05T08:51:25 *** cryptapus is now known as cryptapus_afk 82 2017-06-05T08:54:36 *** d_t has quit IRC 83 2017-06-05T08:57:19 <bitcoin-git> [bitcoin] Flowdalic opened pull request #10529: Improve bitcoind systemd service file (master...systemd-service) https://github.com/bitcoin/bitcoin/pull/10529 84 2017-06-05T09:00:45 *** Guest34260 has joined #bitcoin-core-dev 85 2017-06-05T09:01:03 <_flow_> #10529 was me, in case you have any questions 86 2017-06-05T09:01:04 <gribble> https://github.com/bitcoin/bitcoin/issues/10529 | Improve bitcoind systemd service file by Flowdalic Â· Pull Request #10529 Â· bitcoin/bitcoin Â· GitHub 87 2017-06-05T09:24:12 *** Guest34260 has quit IRC 88 2017-06-05T09:48:56 *** spudowiar has joined #bitcoin-core-dev 89 2017-06-05T09:49:08 <spudowiar> NicolasDorier: I meant for the communication with the hardware wallet 90 2017-06-05T09:49:14 <spudowiar> The hardware wallet plugins use JSON-RPC 91 2017-06-05T09:49:21 <spudowiar> https://github.com/saleemrashid/bitcoin/blob/hardware-wallet/src/wallet/wallet.cpp#L1526 92 2017-06-05T09:49:46 <spudowiar> I called the method signrawtransaction, but it's not compatible (as it sends hdKeypaths instead of privkeys) 93 2017-06-05T09:50:14 <spudowiar> Also, I want to add an identifier, so we can support multiple wallets attached to the computer 94 2017-06-05T09:50:18 <spudowiar> e.g. a serial number 95 2017-06-05T09:54:25 <bitcoin-git> [bitcoin] pavlosantoniou opened pull request #10530: Fix possibly unsafe accesses of array in class base_uint<BITS>. (master...master) https://github.com/bitcoin/bitcoin/pull/10530 96 2017-06-05T09:54:47 <spudowiar> I'm also going to add a listhwwdevices which could be used on the first run to list all the devices so Bitcoin Core can ask the user to select one 97 2017-06-05T09:54:57 <spudowiar> And a gethwwinfo to get the xpub for the selected device 98 2017-06-05T09:55:29 <spudowiar> Then on first run, Bitcoin Core could allow you to select a vendor and then a device 99 2017-06-05T09:55:33 <spudowiar> And store this in the wallet 100 2017-06-05T10:07:31 *** john1 has joined #bitcoin-core-dev 101 2017-06-05T10:16:45 *** Guyver2 has joined #bitcoin-core-dev 102 2017-06-05T10:23:34 *** RubenSomsen has joined #bitcoin-core-dev 103 2017-06-05T10:28:15 *** rafalcpp has joined #bitcoin-core-dev 104 2017-06-05T10:33:33 *** riemann has joined #bitcoin-core-dev 105 2017-06-05T10:42:52 *** spudowiar has quit IRC 106 2017-06-05T10:51:05 *** goatturneer has joined #bitcoin-core-dev 107 2017-06-05T10:54:53 *** beatrootfarmer has quit IRC 108 2017-06-05T10:58:19 *** tiagotrs has joined #bitcoin-core-dev 109 2017-06-05T11:02:04 *** beatrootfarmer has joined #bitcoin-core-dev 110 2017-06-05T11:05:29 *** goatturneer has quit IRC 111 2017-06-05T11:06:49 *** SopaXorzTaker has joined #bitcoin-core-dev 112 2017-06-05T11:15:29 *** laurentmt has joined #bitcoin-core-dev 113 2017-06-05T11:16:13 *** laurentmt has quit IRC 114 2017-06-05T11:34:46 *** goatturneer has joined #bitcoin-core-dev 115 2017-06-05T11:37:01 <NicolasDorier> spudowiar: for multiple wallet, you don't need serial number. There is the multi wallet PR which use RPC User/Password as a way to selection a wallet 116 2017-06-05T11:37:27 <NicolasDorier> and the wallet already know the keypath given the address 117 2017-06-05T11:37:36 <NicolasDorier> so you don't need to pass it to the RPC method 118 2017-06-05T11:38:51 *** beatrootfarmer has quit IRC 119 2017-06-05T11:42:46 *** timothy has quit IRC 120 2017-06-05T11:46:54 *** timothy has joined #bitcoin-core-dev 121 2017-06-05T11:50:11 *** timothy has quit IRC 122 2017-06-05T11:50:32 *** goatturner has joined #bitcoin-core-dev 123 2017-06-05T11:54:15 *** goatturneer has quit IRC 124 2017-06-05T11:57:34 *** RubenSomsen has quit IRC 125 2017-06-05T12:01:47 *** spudowiar has joined #bitcoin-core-dev 126 2017-06-05T12:02:00 <spudowiar> NicolasDorier: How does the wallet plugin know which hardware device to connect to? 127 2017-06-05T12:02:31 <spudowiar> NicolasDorier: You might be misunderstanding 128 2017-06-05T12:02:32 <spudowiar> I'm not talking about the RPC interface Bitcoin Core normally uses 129 2017-06-05T12:02:47 <spudowiar> I'm talking about JSON-RPC that is sent over stdin to the hardware wallet plugin 130 2017-06-05T12:03:01 <NicolasDorier> oh 131 2017-06-05T12:03:11 <NicolasDorier> yes I was misunderstanding 132 2017-06-05T12:03:28 <spudowiar> Check out https://github.com/saleemrashid/bitcoin/blob/hardware-wallet/src/wallet/wallet.cpp#L1527 133 2017-06-05T12:03:31 <NicolasDorier> so it will be possible to use BTC signrawtransaction as if it was keys in software ? 134 2017-06-05T12:03:38 <spudowiar> Oh yeah 135 2017-06-05T12:03:48 <spudowiar> The normal RPC interface won't change 136 2017-06-05T12:03:56 <NicolasDorier> ha awesome 137 2017-06-05T12:03:59 <spudowiar> Although I might make it compatible as a hardware wallet plugin 138 2017-06-05T12:04:03 <NicolasDorier> that is what I meant 139 2017-06-05T12:04:06 <spudowiar> So you could connect one instance to another :) 140 2017-06-05T12:04:13 <spudowiar> And it thinks the other instance is a hardware wallet :) 141 2017-06-05T12:05:36 <NicolasDorier> spudowiar: very cool. Would you mind once you tested a bit to ACK my PR about hdwatchonly ? This would make your PR a bit smaller if you plan to include what you are doing into core 142 2017-06-05T12:06:26 <NicolasDorier> my PR is hanging around for a while, I am using it in production. But was lacking reviews 143 2017-06-05T12:06:27 <spudowiar> If you check that branch, the first few commits are your PR 144 2017-06-05T12:06:42 <spudowiar> I think I did ACK it, if not I'll do that when I can 145 2017-06-05T12:07:31 <NicolasDorier> yes I saw. Just a way to make your PR smaller, so it will be easier to merge yours later 146 2017-06-05T12:10:06 <spudowiar> The other PR we need is making signing asynchronous in Bitcoin Qt 147 2017-06-05T12:10:35 <spudowiar> At the moment, while it is talking to the hardware wallet the GUI freezes 148 2017-06-05T12:10:53 <spudowiar> Because the io_service is running on the UI thread 149 2017-06-05T12:12:59 <NicolasDorier> rebasing my PR. 150 2017-06-05T12:13:22 <spudowiar> Thanks, I'll ACK and rebase ASAP 151 2017-06-05T12:13:34 <NicolasDorier> actually you already rebased right ? 152 2017-06-05T12:13:45 <NicolasDorier> I can just take your branch up to my commits I think 153 2017-06-05T12:14:50 <spudowiar> Yeah, sure but you might want to check (I had to make a few changes for merge conflicts) 154 2017-06-05T12:15:10 <NicolasDorier> ok 155 2017-06-05T12:15:34 <spudowiar> A three way diff should let you see what I had to change 156 2017-06-05T12:27:40 <spudowiar> Do you think I should add a scan in the PATH for plugins? Or a scan in a folder? 157 2017-06-05T12:27:48 <spudowiar> e.g. bitcoin-hww-* 158 2017-06-05T12:28:12 <spudowiar> Then, on startup, the GUI wallet could display all the vendors 159 2017-06-05T12:36:22 *** RubenSomsen has joined #bitcoin-core-dev 160 2017-06-05T12:39:13 *** timothy has joined #bitcoin-core-dev 161 2017-06-05T12:45:17 <bitcoin-git> [bitcoin] colgreen opened pull request #10531: Increased startup timeout. (master...patch-1) https://github.com/bitcoin/bitcoin/pull/10531 162 2017-06-05T13:01:57 *** spudowiar has quit IRC 163 2017-06-05T13:07:05 <NicolasDorier> spudowiar: unsure, I like your current approach to just delegate to a child process specified by command line. 164 2017-06-05T13:09:10 *** Dyaheon has quit IRC 165 2017-06-05T13:10:51 *** Dyaheon has joined #bitcoin-core-dev 166 2017-06-05T13:12:37 <wumpus> I also like the external process approach 167 2017-06-05T13:13:26 <wumpus> loading plugins into bitcoind/-qt 's address space seem like the wrong way, for me, and is hard harmonize with the static linking we do for distributed executables 168 2017-06-05T13:13:38 *** Taek42 is now known as Taek 169 2017-06-05T13:14:38 *** AaronvanW has quit IRC 170 2017-06-05T13:15:00 *** AaronvanW has joined #bitcoin-core-dev 171 2017-06-05T13:15:26 *** Yogaqueef has joined #bitcoin-core-dev 172 2017-06-05T13:18:05 *** goatturneer has joined #bitcoin-core-dev 173 2017-06-05T13:21:23 *** bsm1175321 has joined #bitcoin-core-dev 174 2017-06-05T13:21:59 *** goatturner has quit IRC 175 2017-06-05T13:24:38 *** JackH has joined #bitcoin-core-dev 176 2017-06-05T13:30:36 *** bsm1175321 has quit IRC 177 2017-06-05T13:32:51 *** bsm1175321 has joined #bitcoin-core-dev 178 2017-06-05T13:34:10 *** bsm1175321 has quit IRC 179 2017-06-05T13:40:39 <bitcoin-git> [bitcoin] laanwj pushed 3 new commits to master: https://github.com/bitcoin/bitcoin/compare/400fdd08cc95...296928eb38a4 180 2017-06-05T13:40:40 <bitcoin-git> bitcoin/master a7e3c28 Wladimir J. van der Laan: rpc: Add listen address to incoming connections in `getpeerinfo`... 181 2017-06-05T13:40:41 <bitcoin-git> bitcoin/master 3457331 Wladimir J. van der Laan: test: Add test for `getpeerinfo` `bindaddr` field 182 2017-06-05T13:40:41 <bitcoin-git> bitcoin/master 296928e Wladimir J. van der Laan: Merge #10478: rpc: Add listen address to incoming connections in `getpeerinfo`... 183 2017-06-05T13:41:19 <bitcoin-git> [bitcoin] laanwj closed pull request #10478: rpc: Add listen address to incoming connections in `getpeerinfo` (master...2017_05_peer_listenaddr) https://github.com/bitcoin/bitcoin/pull/10478 184 2017-06-05T13:47:06 *** bsm1175321 has joined #bitcoin-core-dev 185 2017-06-05T13:50:36 *** d_t has joined #bitcoin-core-dev 186 2017-06-05T13:53:40 *** chjj has quit IRC 187 2017-06-05T13:58:01 *** beatrootfarmer has joined #bitcoin-core-dev 188 2017-06-05T14:01:32 *** goatturneer has quit IRC 189 2017-06-05T14:07:39 *** chjj has joined #bitcoin-core-dev 190 2017-06-05T14:10:25 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/296928eb38a4...e103b3ff1e4f 191 2017-06-05T14:10:25 <bitcoin-git> bitcoin/master 9aa215b Pieter Wuille: Bugfixes: missing == 0 after randrange 192 2017-06-05T14:10:26 <bitcoin-git> bitcoin/master e103b3f Wladimir J. van der Laan: Merge #10514: Bugfix: missing == 0 after randrange... 193 2017-06-05T14:10:59 <bitcoin-git> [bitcoin] laanwj closed pull request #10514: Bugfix: missing == 0 after randrange (master...fixtests) https://github.com/bitcoin/bitcoin/pull/10514 194 2017-06-05T14:15:54 *** Guyver2 has quit IRC 195 2017-06-05T14:18:48 *** wangchun has quit IRC 196 2017-06-05T14:20:33 *** yellowsphere_ has joined #bitcoin-core-dev 197 2017-06-05T14:24:48 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/e103b3ff1e4f...cec9e1ea6174 198 2017-06-05T14:24:49 <bitcoin-git> bitcoin/master 8d4f401 Alex Morcos: Fix timestamp in fee estimate debug message 199 2017-06-05T14:24:49 <bitcoin-git> bitcoin/master cec9e1e Wladimir J. van der Laan: Merge #10422: Fix timestamp in fee estimate debug message... 200 2017-06-05T14:25:26 <bitcoin-git> [bitcoin] laanwj closed pull request #10422: Fix timestamp in fee estimate debug message (master...fixtimeunits) https://github.com/bitcoin/bitcoin/pull/10422 201 2017-06-05T14:25:26 *** wangchun has joined #bitcoin-core-dev 202 2017-06-05T14:28:03 <bitcoin-git> [bitcoin] MarcoFalke pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/cec9e1ea6174...c871f323b418 203 2017-06-05T14:28:04 <bitcoin-git> bitcoin/master cc36b5e Jimmy Song: [test] Add test for getchaintxstats 204 2017-06-05T14:28:05 <bitcoin-git> bitcoin/master c871f32 MarcoFalke: Merge #10515: [test] Add test for getchaintxstats... 205 2017-06-05T14:28:39 <bitcoin-git> [bitcoin] MarcoFalke closed pull request #10515: [test] Add test for getchaintxstats (master...test_getchaintxstats) https://github.com/bitcoin/bitcoin/pull/10515 206 2017-06-05T14:34:26 *** Chris_Stewart_5 has joined #bitcoin-core-dev 207 2017-06-05T14:38:28 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/c871f323b418...0747d3349817 208 2017-06-05T14:38:29 <bitcoin-git> bitcoin/master 88b8f0b Russell Yanofsky: Simplify feebumper minimum fee code slightly... 209 2017-06-05T14:38:29 <bitcoin-git> bitcoin/master 0747d33 Wladimir J. van der Laan: Merge #10455: Simplify feebumper minimum fee code slightly... 210 2017-06-05T14:39:09 <bitcoin-git> [bitcoin] laanwj closed pull request #10455: Simplify feebumper minimum fee code slightly (master...pr/bumpmin) https://github.com/bitcoin/bitcoin/pull/10455 211 2017-06-05T14:40:45 *** talmai has joined #bitcoin-core-dev 212 2017-06-05T14:41:24 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/0747d3349817...08d0390a5fea 213 2017-06-05T14:41:24 <bitcoin-git> bitcoin/master 4d2d604 Russell Yanofsky: Fix importmulti failure to return rescan errors... 214 2017-06-05T14:41:25 <bitcoin-git> bitcoin/master 08d0390 Wladimir J. van der Laan: Merge #10403: Fix importmulti failure to return rescan errors... 215 2017-06-05T14:41:56 <bitcoin-git> [bitcoin] laanwj closed pull request #10403: Fix importmulti failure to return rescan errors (master...pr/scansame) https://github.com/bitcoin/bitcoin/pull/10403 216 2017-06-05T14:44:57 *** talmai has quit IRC 217 2017-06-05T14:47:51 *** chjj has quit IRC 218 2017-06-05T14:47:58 *** tiagotrs has quit IRC 219 2017-06-05T14:55:07 *** JackH has quit IRC 220 2017-06-05T15:00:17 *** riemann has quit IRC 221 2017-06-05T15:00:21 <bitcoin-git> [bitcoin] laanwj pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/08d0390a5fea...e654d61d94ee 222 2017-06-05T15:00:22 <bitcoin-git> bitcoin/master cf390df Cory Fields: build: silence gcc7's implicit fallthrough warning... 223 2017-06-05T15:00:22 <bitcoin-git> bitcoin/master e654d61 Wladimir J. van der Laan: Merge #10489: build: silence gcc7's implicit fallthrough warning... 224 2017-06-05T15:00:56 <bitcoin-git> [bitcoin] laanwj closed pull request #10489: build: silence gcc7's implicit fallthrough warning (master...wimplicit-fallthrough) https://github.com/bitcoin/bitcoin/pull/10489 225 2017-06-05T15:02:30 *** RubenSomsen has quit IRC 226 2017-06-05T15:12:29 *** RubenSomsen has joined #bitcoin-core-dev 227 2017-06-05T15:15:05 *** arubi has quit IRC 228 2017-06-05T15:15:33 *** str4d has quit IRC 229 2017-06-05T15:15:39 *** arubi has joined #bitcoin-core-dev 230 2017-06-05T15:18:24 *** goatturneer has joined #bitcoin-core-dev 231 2017-06-05T15:19:59 *** bsm1175321 has quit IRC 232 2017-06-05T15:20:16 *** ivo___ has joined #bitcoin-core-dev 233 2017-06-05T15:21:44 <ivo___> \disconnect 234 2017-06-05T15:21:57 *** ivo___ has quit IRC 235 2017-06-05T15:22:23 *** beatrootfarmer has quit IRC 236 2017-06-05T15:22:30 *** bsm1175321 has joined #bitcoin-core-dev 237 2017-06-05T15:34:41 *** emzy has quit IRC 238 2017-06-05T15:35:56 *** emzy has joined #bitcoin-core-dev 239 2017-06-05T15:49:45 *** abpa has joined #bitcoin-core-dev 240 2017-06-05T15:50:05 <bitcoin-git> [bitcoin] MarcoFalke pushed 2 new commits to master: https://github.com/bitcoin/bitcoin/compare/e654d61d94ee...bea5b00cfe95 241 2017-06-05T15:50:05 <bitcoin-git> bitcoin/master e4bc19f Russell Yanofsky: Remove xvfb configuration from travis... 242 2017-06-05T15:50:06 <bitcoin-git> bitcoin/master bea5b00 MarcoFalke: Merge #10509: Remove xvfb configuration from travis... 243 2017-06-05T15:50:40 <bitcoin-git> [bitcoin] MarcoFalke closed pull request #10509: Remove xvfb configuration from travis (master...pr/rmfb) https://github.com/bitcoin/bitcoin/pull/10509 244 2017-06-05T15:59:38 <bitcoin-git> [bitcoin] laanwj closed pull request #9745: [RPC] Getting confirmations command (master...add-getconfirmations-to-rpc) https://github.com/bitcoin/bitcoin/pull/9745 245 2017-06-05T16:01:53 *** emzy has quit IRC 246 2017-06-05T16:01:54 *** emzy has joined #bitcoin-core-dev 247 2017-06-05T16:06:11 *** laurentmt has joined #bitcoin-core-dev 248 2017-06-05T16:06:55 <bitcoin-git> [bitcoin] laanwj pushed 3 new commits to master: https://github.com/bitcoin/bitcoin/compare/bea5b00cfe95...9fec4da0bec9 249 2017-06-05T16:06:56 <bitcoin-git> bitcoin/master 9f82134 aideca: Add friendly output to dumpwallet refs #9564 250 2017-06-05T16:06:56 <bitcoin-git> bitcoin/master 164019d aideca: Add dumpwallet output test 251 2017-06-05T16:06:57 <bitcoin-git> bitcoin/master 9fec4da Wladimir J. van der Laan: Merge #9740: Add friendly output to dumpwallet... 252 2017-06-05T16:07:15 <bitcoin-git> [bitcoin] laanwj closed pull request #9740: Add friendly output to dumpwallet (master...dumpwallet) https://github.com/bitcoin/bitcoin/pull/9740 253 2017-06-05T16:08:21 *** beatrootfarmer has joined #bitcoin-core-dev 254 2017-06-05T16:09:17 *** spudowiar has joined #bitcoin-core-dev 255 2017-06-05T16:09:39 <spudowiar> NicolasDorier, wumpus: I didn't mean changing it from an external command 256 2017-06-05T16:09:49 <spudowiar> Just a method of detecting the available external commands 257 2017-06-05T16:10:09 <spudowiar> But I could add it to the wallet file or something 258 2017-06-05T16:10:23 <spudowiar> And provide an option at startup of Bitcoin-Qt to enter the command 259 2017-06-05T16:12:19 *** goatturneer has quit IRC 260 2017-06-05T16:13:43 *** Dyaheon has quit IRC 261 2017-06-05T16:15:35 *** Dyaheon has joined #bitcoin-core-dev 262 2017-06-05T16:19:31 *** nsh- is now known as nsh 263 2017-06-05T16:20:34 *** spudowiar has quit IRC 264 2017-06-05T16:20:41 *** spudowiar has joined #bitcoin-core-dev 265 2017-06-05T16:24:37 <bitcoin-git> [bitcoin] earonesty opened pull request #10532: -bip148 option (master...bip148) https://github.com/bitcoin/bitcoin/pull/10532 266 2017-06-05T16:39:18 *** laurentmt has quit IRC 267 2017-06-05T16:40:48 <btcdrak> wumpus: could you restart the two failing jobs here https://travis-ci.org/bitcoin/bitcoin/builds/238932256 268 2017-06-05T16:52:48 *** john1 has quit IRC 269 2017-06-05T16:56:59 *** laurentmt has joined #bitcoin-core-dev 270 2017-06-05T16:58:46 <spudowiar> sipa__: Is there some good reading about the polymod function Bech32 uses? 271 2017-06-05T17:01:38 *** timothy has quit IRC 272 2017-06-05T17:10:35 <BlueMatt> ;;later tell sipa sipa__ is it just my C++11-lack-of-sanity, or am I suppose to find https://github.com/bitcoin/bitcoin/blob/master/src/coins.cpp#L99 gross as fuck? std::move(it->second.coin) then a few lines later it->second.coin.function_call()? 273 2017-06-05T17:10:35 <gribble> The operation succeeded. 274 2017-06-05T17:13:07 <cfields> BlueMatt: after a move, contents are usually undefined, but valid 275 2017-06-05T17:13:17 <cfields> BlueMatt: a clear after a move makes them defined and valid :) 276 2017-06-05T17:13:26 <BlueMatt> cfields: ugh, that just feels super gros 277 2017-06-05T17:13:27 <BlueMatt> s 278 2017-06-05T17:13:50 <BlueMatt> since that line depends on the exact behavior of CScript::Clear(), prevector, and CTxOut 279 2017-06-05T17:15:01 <cfields> BlueMatt: the alternative would be defining an explicit move ctor which does the clear and leaves it in a valid state 280 2017-06-05T17:15:12 <cfields> (or, i suppose, you'd rather just have it not used after the move) 281 2017-06-05T17:15:35 <BlueMatt> yea, I mean I generally read std::move(X) as "X will not be touched again" 282 2017-06-05T17:15:44 <BlueMatt> maybe thats just my naivete on C++11 283 2017-06-05T17:15:53 <cfields> BlueMatt: well, that's not really feasible with a container :\ 284 2017-06-05T17:16:20 <cfields> std::move(foo), foo still exists just fine. 285 2017-06-05T17:16:45 <cfields> In many cases, this one included i think, it's the only way to avoid incurring the copy when moving the contents of a member 286 2017-06-05T17:17:51 <BlueMatt> cfields: well obv destructor must run fine after std::move(), but to me I read that as, "I wont touch the 0th element of foo anymore after this, except to destruct" 287 2017-06-05T17:18:24 <sipa__> BlueMatt: typically the definition os that you're allowed to do anything that foesn't depend on otd prior state 288 2017-06-05T17:18:32 <sipa__> so you can assign, clear, ... 289 2017-06-05T17:19:00 <sipa__> in a vector, a move-from + .clear() is well specified 290 2017-06-05T17:19:19 <sipa__> doesn't 291 2017-06-05T17:20:27 *** laurentmt has quit IRC 292 2017-06-05T17:21:26 <MarcoFalke> btcdrak: done. Looks like they hit the 50 min timeout :( 293 2017-06-05T17:25:02 <sipa__> BlueMatt: https://stackoverflow.com/a/7028318 294 2017-06-05T17:25:13 *** sipa__ is now known as sipa 295 2017-06-05T17:25:33 <BlueMatt> sipa__: yea, but coin.Clear() is super strange, it calls CScript::clear() which doesnot prevector::clear(), instead it prevector().swap(*this), essentially 296 2017-06-05T17:25:35 <BlueMatt> which is...wat 297 2017-06-05T17:25:54 <BlueMatt> sipa: can we make it it->second.coin = CCoin() instead? 298 2017-06-05T17:26:06 <BlueMatt> should be +/- identical performance, but without feeling so yucky 299 2017-06-05T17:26:36 *** sipa has quit IRC 300 2017-06-05T17:26:36 *** sipa has joined #bitcoin-core-dev 301 2017-06-05T17:27:12 <sipa> i don't see the difference 302 2017-06-05T17:27:31 <sipa> both the Clear method and assignment operator can be user defined 303 2017-06-05T17:28:01 <BlueMatt> sure, swap just feels worse to me...I'll shut up if no one else thinks it feels strange, though :) 304 2017-06-05T17:29:44 <cfields> BlueMatt: the swap idiom comes from pre-c++11 when there was no shrink_to_fit(). 305 2017-06-05T17:30:15 <cfields> imo the clear shouldn't be forcing mem dealloc, it should just be clearing, and we should be using shrink_to_fit() more liberally at higher layers. 306 2017-06-05T17:30:30 <sipa> right 307 2017-06-05T17:30:43 <BlueMatt> yes, prevector clear() doesnt, script clear() does 308 2017-06-05T17:31:10 <BlueMatt> could make script more vector-like by pushing it up to CCoin Clear() 309 2017-06-05T17:31:16 <BlueMatt> or CTxOut 310 2017-06-05T17:31:30 <sipa> script clear could be changed into prevector clear + shrink to fit 311 2017-06-05T17:31:52 <sipa> though i think prevector clear automatocally shrinks 312 2017-06-05T17:31:59 <BlueMatt> std::move() + clear() + shrink_to_fit()? 313 2017-06-05T17:32:03 <BlueMatt> i do not believe so 314 2017-06-05T17:32:23 <cfields> oh right. prevector essentially has no shrink_to_fit because it never over-allocs 315 2017-06-05T17:32:44 <BlueMatt> ohoh, true 316 2017-06-05T17:33:39 <BlueMatt> ehh, no, wait, i dont think so? 317 2017-06-05T17:34:48 <BlueMatt> erase(it, it) doesnt do any shrinking, i believe 318 2017-06-05T17:37:28 <sipa> when usimg direct allocation, the capacity is always equal to the object size 319 2017-06-05T17:37:46 <sipa> when using indirect allocation, the capacity is variable 320 2017-06-05T17:37:52 <BlueMatt> yes, but if its grown, clear() doesnt shrink_to_fit 321 2017-06-05T17:38:22 <sipa> so resizing to something less than the inline capacity always has a deallocation as a result 322 2017-06-05T17:38:26 <sipa> or at least should have 323 2017-06-05T17:38:29 <MarcoFalke> cfields: I had the impression that the wallet-block_tip race is still a thing. Otherwise we could get rid of waitforblockheight as well 324 2017-06-05T17:38:34 <sipa> as there is no way to represent the alternative 325 2017-06-05T17:40:15 <cfields> MarcoFalke: yes, that's right. But we don't have any tests that rely on the other 2 things being atomic. 326 2017-06-05T17:40:16 <sipa> i'm now questioning whether the prevector code is correct... 327 2017-06-05T17:40:27 <cfields> sipa: i was just thinking the same. that seems leaky :\ 328 2017-06-05T17:41:26 <BlueMatt> sipa: yes, there is a bug on erase(begin(), end()), but not on clear()/resize(0) 329 2017-06-05T17:42:03 *** laurentmt has joined #bitcoin-core-dev 330 2017-06-05T17:42:16 <sipa> clear() just calls erase all the thing 331 2017-06-05T17:42:27 <sipa> erase has no change capacity call at all 332 2017-06-05T17:48:15 *** chjj has joined #bitcoin-core-dev 333 2017-06-05T17:51:16 *** john1 has joined #bitcoin-core-dev 334 2017-06-05T17:52:12 *** ula has joined #bitcoin-core-dev 335 2017-06-05T17:52:25 *** chjj has quit IRC 336 2017-06-05T17:52:46 *** chjj has joined #bitcoin-core-dev 337 2017-06-05T17:56:05 *** QBcrusher has joined #bitcoin-core-dev 338 2017-06-05T17:57:15 *** AaronvanW has quit IRC 339 2017-06-05T17:57:49 *** yellowsphere_ has quit IRC 340 2017-06-05T17:57:52 *** AaronvanW has joined #bitcoin-core-dev 341 2017-06-05T17:58:27 *** QBcrusher_ has quit IRC 342 2017-06-05T18:07:52 *** owowo has quit IRC 343 2017-06-05T18:09:49 *** laurentmt has quit IRC 344 2017-06-05T18:16:22 <jtimon> https://github.com/bitcoin/bitcoin/pull/10502 was created 4 days ago and laready needed rebase twice 345 2017-06-05T18:16:28 <jtimon> already 346 2017-06-05T18:17:10 <sipa> jtimon: thankfully, it's a scripted diff mostly, so should be easy 347 2017-06-05T18:17:23 *** Dyaheon has quit IRC 348 2017-06-05T18:17:24 <jtimon> it is disruptive (+296 â300 at this point), so it is likely to require more 349 2017-06-05T18:17:45 <jtimon> sipa: yeah, I think we can get it reviewed fast too 350 2017-06-05T18:20:47 *** Dyaheon has joined #bitcoin-core-dev 351 2017-06-05T18:21:10 <jtimon> #10463 has some utACKs already and is easy to review as well 352 2017-06-05T18:21:11 <gribble> https://github.com/bitcoin/bitcoin/issues/10463 | Names: BIP9 vs versionbits by jtimon Â· Pull Request #10463 Â· bitcoin/bitcoin Â· GitHub 353 2017-06-05T18:22:32 <bitcoin-git> [bitcoin] achow101 opened pull request #10533: [tests] Use cookie auth instead of rpcuser and rpcpassword (master...tests-use-cookie-auth) https://github.com/bitcoin/bitcoin/pull/10533 354 2017-06-05T18:22:48 *** john1 has quit IRC 355 2017-06-05T18:24:34 *** JackH has joined #bitcoin-core-dev 356 2017-06-05T18:34:09 *** chjj has quit IRC 357 2017-06-05T18:38:02 *** Chris_Stewart_5 has quit IRC 358 2017-06-05T18:40:15 *** harrymm has quit IRC 359 2017-06-05T18:42:39 *** talmai has joined #bitcoin-core-dev 360 2017-06-05T18:46:48 *** chjj has joined #bitcoin-core-dev 361 2017-06-05T18:47:30 *** Chris_Stewart_5 has joined #bitcoin-core-dev 362 2017-06-05T18:49:11 *** john1 has joined #bitcoin-core-dev 363 2017-06-05T18:56:05 *** harrymm has joined #bitcoin-core-dev 364 2017-06-05T19:01:55 *** JackH has quit IRC 365 2017-06-05T19:06:53 <BlueMatt> sipa: ah, indeed, no bug in prevector, I was misreading the meaning of _size 366 2017-06-05T19:06:54 <jtimon> sipa: regarding https://github.com/bitcoin/bitcoin/pull/10502#issuecomment-306275178 perhaps you can help me find out what's wrong with https://github.com/bitcoin/bitcoin/pull/10193/commits/0f151d6a1b5b048ca78879e367f7a18f91126066#diff-c295474658e4ef7a9f5324199375edaf 367 2017-06-05T19:07:33 <sipa> BlueMatt: oh, good 368 2017-06-05T19:10:44 <jtimon> well, except for the reverse lop thing breaking the unittests... 369 2017-06-05T19:11:01 <jtimon> loop 370 2017-06-05T19:12:38 <sipa> ah i see 371 2017-06-05T19:13:11 <jtimon> I haven't been able to understand what's going on there 372 2017-06-05T19:13:11 <sipa> https://status.github.com/ 373 2017-06-05T19:13:51 <jtimon> oh, http://downforeveryoneorjustme.com/https://github.com/bitcoin/bitcoin/ said it was just me 374 2017-06-05T19:14:46 <sipa> it's up again 375 2017-06-05T19:15:02 <jtimon> nice, thanks 376 2017-06-05T19:18:14 <bitcoin-git> [bitcoin] sipa opened pull request #10534: Clarify prevector::erase and avoid swap-to-clear (master...clarify_erase) https://github.com/bitcoin/bitcoin/pull/10534 377 2017-06-05T19:19:55 *** chjj has quit IRC 378 2017-06-05T19:23:10 *** RubenSomsen has quit IRC 379 2017-06-05T19:25:18 *** JackH has joined #bitcoin-core-dev 380 2017-06-05T19:28:29 <cfields> jtimon: i think your expression goes out of scope after the first iteration 381 2017-06-05T19:31:11 <jtimon> cfields: mhmm, I don't follow 382 2017-06-05T19:31:40 *** goatturneer has joined #bitcoin-core-dev 383 2017-06-05T19:33:56 <cfields> jtimon: try: https://pastebin.com/raw/USdQhE36 384 2017-06-05T19:34:12 *** chjj has joined #bitcoin-core-dev 385 2017-06-05T19:35:26 *** beatrootfarmer has quit IRC 386 2017-06-05T19:42:48 *** cryptapus_afk has quit IRC 387 2017-06-05T19:45:23 <spudowiar> I added some stuff to the Intro Qt form for hardware wallet support... Qt Designer made me cry :( 388 2017-06-05T19:46:39 <jtimon> cfields: awesome, I'll have a look, thanks 389 2017-06-05T19:49:00 <BlueMatt> sipa: would you be upset if I PRd making AccessCoin return a CoinAccessor which just asserts if you modify the coinscache while holding a reference? I know its not illegal 390 2017-06-05T19:49:15 <BlueMatt> cause map, but I'm afraid jeremy or I will at some point suggest using a different map 391 2017-06-05T19:49:18 <BlueMatt> and then introduce a bug 392 2017-06-05T19:49:29 <BlueMatt> should be +/- free performance wise 393 2017-06-05T19:49:54 <sipa> BlueMatt: that seems more something for a sanity checker 394 2017-06-05T19:50:06 <sipa> why specifically for CCoinsViewCache and not every other map? 395 2017-06-05T19:50:56 <BlueMatt> sipa: cause I have a strong suspicion over the next release or two someone is gonna realize std::unordered_map is not the ideal map and decide to suggest a hand-written one 396 2017-06-05T19:51:10 *** spudowiar has quit IRC 397 2017-06-05T19:51:11 <sipa> yes, so? 398 2017-06-05T19:51:29 <BlueMatt> and then suddenly using the std sanity-check mode wont help? 399 2017-06-05T19:51:55 <sipa> yes, so? 400 2017-06-05T19:52:04 <sipa> i mean things like valgrind exist to find bugs like that 401 2017-06-05T19:52:39 <sipa> i plan to work on replacing the std::unordered_map myself there, btw 402 2017-06-05T19:52:42 <BlueMatt> cause I'd be not at all surprised if it were very difficult to tickle that issue during valgrind 403 2017-06-05T19:52:45 <jeremyrubin> Can also use a weak reference or something 404 2017-06-05T19:52:45 <BlueMatt> lol, figured someone would 405 2017-06-05T19:53:11 <sipa> i don't understand why you're specifically worried about this map 406 2017-06-05T19:53:11 <BlueMatt> jeremyrubin: that may be incompatible with a smarter map, sadly :( 407 2017-06-05T19:53:28 <jeremyrubin> Can just add an "updates" counter to the map 408 2017-06-05T19:53:49 <jeremyrubin> and the weak ref is invalid if counter != weakref.counter 409 2017-06-05T19:53:50 <sipa> it's a big and important one, but hardly the only complex data structure around 410 2017-06-05T19:54:16 <gmaxwell> BlueMatt: please rebase #10192; review #10148 411 2017-06-05T19:54:19 <BlueMatt> sipa: well we hold references into that map in rpc functions...I'm not aware of any other map where we do anything like that? 412 2017-06-05T19:54:20 <gribble> https://github.com/bitcoin/bitcoin/issues/10192 | Cache full script execution results in addition to signatures by TheBlueMatt Â· Pull Request #10192 Â· bitcoin/bitcoin Â· GitHub 413 2017-06-05T19:54:21 <gribble> https://github.com/bitcoin/bitcoin/issues/10148 | [WIP] Use non-atomic flushing with block replay by sipa Â· Pull Request #10148 Â· bitcoin/bitcoin Â· GitHub 414 2017-06-05T19:54:31 <sipa> BlueMatt: example? 415 2017-06-05T19:54:34 <BlueMatt> gmaxwell: yesyes, I'm working on a few small things i wanted in per-utxo first 416 2017-06-05T19:54:34 <jeremyrubin> anyways, I'm generally sympathetic to Matt's request; long held iterators to maps -> bugs 417 2017-06-05T19:54:46 <BlueMatt> well, luckily for now we have no "long held" ones 418 2017-06-05T19:54:50 <BlueMatt> but we do have ones in RPC 419 2017-06-05T19:54:52 <sipa> jtimon: not long held iterators; references to elements 420 2017-06-05T19:54:58 <sipa> eh, jeremyrubin 421 2017-06-05T19:55:04 <sipa> iterators are encapsulated 422 2017-06-05T19:55:08 <BlueMatt> sipa: AccessCoin is called in rpc/rawtransaction 423 2017-06-05T19:55:10 <BlueMatt> in several places 424 2017-06-05T19:55:33 <sipa> in many places we also hold an iterator that comes out of mapBlockIndex.find 425 2017-06-05T19:55:55 <BlueMatt> sipa: I'm sure this comes back to your "objects are transparent holders of data" vs my "objects should be opaque, where possible" 426 2017-06-05T19:56:01 <gmaxwell> BlueMatt: k. other people's review is blocked on that rebase. 427 2017-06-05T19:56:24 <BlueMatt> gmaxwell: hey, you're the one who got me to focus on fibre for two weeks :p 428 2017-06-05T19:56:32 <BlueMatt> gmaxwell: but will try to circle around today 429 2017-06-05T19:57:16 <BlueMatt> sipa: hmm, somehow i feel better about that cause its clear its a map, unlike CCoinsViewCache, where its an encapsulated datastructure that is actually private 430 2017-06-05T19:57:46 <sipa> BlueMatt: it does return a reference, which is always a worry 431 2017-06-05T19:57:57 <gmaxwell> BlueMatt: I know, I know. 432 2017-06-05T19:58:08 <jeremyrubin> BlueMatt: at a higher level, I see what you're trying to do, trying to get me to write a better map and I'm maybe falling for it 433 2017-06-05T19:58:23 <BlueMatt> jeremyrubin: sipa already said he was working on it :p 434 2017-06-05T19:58:31 <sipa> if you're worried about particular call sites, you can always use Coin coin = view.Accessor(outpoint) ... instead of Coin& coin = ... 435 2017-06-05T19:58:49 <BlueMatt> sipa: our disagreement is clearly a disagreement about encapsulation 436 2017-06-05T19:59:13 <sipa> BlueMatt: i'm just so glad i got rid of that CCoinsModifier... and now you want to introduce it again :) 437 2017-06-05T19:59:21 <sipa> the interface got so much easier 438 2017-06-05T19:59:30 <BlueMatt> sipa: CCoinsViewModifier *did* things, there I agree that is gross 439 2017-06-05T19:59:45 <BlueMatt> I'm proposing something who's only value is to assert with litte/no performance impact 440 2017-06-05T20:00:02 <BlueMatt> so as to act as a sanity checker on callsites to a function I cant say I'm sold on 441 2017-06-05T20:00:21 <gmaxwell> BlueMatt: what sanity check will it do? 442 2017-06-05T20:00:44 <sipa> gmaxwell: check that no references are held while a modify happens 443 2017-06-05T20:01:15 <BlueMatt> gmaxwell: technically assert something that is currently allowed, but something I assume will be come unallowed when sipa (or someone) writes a custom map for mapCoins 444 2017-06-05T20:01:39 <gmaxwell> that doesn't sound free to verify. 445 2017-06-05T20:02:03 <sipa> BlueMatt: also, i don't think it's technically illegal to hold a reference to a deleted object 446 2017-06-05T20:02:05 <BlueMatt> hmm? just a boolean/int to set when you AccessCoins and assert() on in function calls 447 2017-06-05T20:02:16 <sipa> only using it is 448 2017-06-05T20:02:22 <jeremyrubin> what if you have a templated thing that doesn't check that property if the type of map used is the current one ;p 449 2017-06-05T20:02:43 <BlueMatt> gmaxwell: compared to looking up in the map it is free :) 450 2017-06-05T20:03:13 <gmaxwell> BlueMatt: not necessarily compared to an alternative data structure (also, I doubt it's free even compared to a map) 451 2017-06-05T20:03:17 <BlueMatt> sipa: sure, but cutting hairs, I'm asking if you're ok adding a class which is more strict than it needs to be, cause its easy to write and, imo, a useful sanity check 452 2017-06-05T20:03:18 <ryanofsky> sipa, not directly related but #9384 also adds back a version of the modifier 453 2017-06-05T20:03:27 <gribble> https://github.com/bitcoin/bitcoin/issues/9384 | CCoinsViewCache code cleanup & deduplication by ryanofsky Â· Pull Request #9384 Â· bitcoin/bitcoin Â· GitHub 454 2017-06-05T20:03:28 *** JackH has quit IRC 455 2017-06-05T20:03:31 <BlueMatt> gmaxwell: assert(nAccessors == 0); ? 456 2017-06-05T20:03:32 <sipa> ryanofsky: yes... but not an exposed one 457 2017-06-05T20:03:53 <ryanofsky> yeah 458 2017-06-05T20:03:58 <BlueMatt> gmaxwell: that sounds like something that the cpu will run without any attached preconditions and will easily run free since you're informing the branch predictor what is unlikely 459 2017-06-05T20:04:19 <sipa> ryanofsky: i like that 9384 reduces logic duplication, i dislike that it does not actually reduce the amount of code :) 460 2017-06-05T20:05:41 <ryanofsky> i think that could be because i wrote it in an intentionally verbose style with lots of comments and named variables 461 2017-06-05T20:06:02 <sipa> ryanofsky: yeah, will review 462 2017-06-05T20:06:07 <sipa> that was just a quick comment 463 2017-06-05T20:07:01 <sipa> BlueMatt: so, an accessor object may have another advantage at some point... if we'd want to make CCoinsViewCache have split storage for txids and CCoins 464 2017-06-05T20:07:04 <sipa> eh, Coins 465 2017-06-05T20:07:26 <sipa> i don't know if that's a good idea or not, but it may reduce memory usage a fair bit (or not) 466 2017-06-05T20:07:37 <BlueMatt> you mean to hold a ptr to uint256? 467 2017-06-05T20:08:08 *** BashCo has quit IRC 468 2017-06-05T20:08:10 <sipa> well the point is that encapsulating it may give freedom to change the representation more liberally 469 2017-06-05T20:08:13 <BlueMatt> sipa: if you really dont want an accessor object, I'ma add a nice long comment to AccessCoin describing the preciditions...which I'm ok with, but not as much a fan 470 2017-06-05T20:08:15 *** BashCo has joined #bitcoin-core-dev 471 2017-06-05T20:08:15 <BlueMatt> hence the q 472 2017-06-05T20:08:19 <BlueMatt> fair 473 2017-06-05T20:08:37 <sipa> so how about we benchmark whether it has an effect? 474 2017-06-05T20:08:46 <BlueMatt> do we have a decent way to benchmark? 475 2017-06-05T20:08:59 <BlueMatt> i mean could reindex 476 2017-06-05T20:09:07 <sipa> reindex chainstate on a -connect=0 node with fixed CPU clockrate 477 2017-06-05T20:09:08 <BlueMatt> but that nearly definitely wont show it 478 2017-06-05T20:09:11 <sipa> works very accurately 479 2017-06-05T20:09:21 <BlueMatt> I so dont want to do that....I can haz ssh? 480 2017-06-05T20:10:19 <sipa> i can set things up 481 2017-06-05T20:22:20 *** harrymm has quit IRC 482 2017-06-05T20:24:28 *** Dyaheon has quit IRC 483 2017-06-05T20:24:56 *** Dyaheon has joined #bitcoin-core-dev 484 2017-06-05T20:29:37 <BlueMatt> sipa: well, went and did it and no clear way to enforce semantics cause you can take reference and then drop the accessor too easily when writing cleints. it is nice cause it forces back to a ptr and restores a bunch of effectively-assert-cause-nullptrs that you removed in per-utxo 485 2017-06-05T20:29:48 <BlueMatt> maybe I'll just re-add asserts and a long comment 486 2017-06-05T20:30:40 <bitcoin-git> [bitcoin] practicalswift closed pull request #10527: Use parentheses to clarify intended precedence when using bitwise operations (master...clarify-precedence) https://github.com/bitcoin/bitcoin/pull/10527 487 2017-06-05T20:35:52 <Chris_Stewart_5> FYI: You can install bitcoind as a dependency now on travis ci, previously it had been black listed. 488 2017-06-05T20:36:16 *** harrymm has joined #bitcoin-core-dev 489 2017-06-05T20:37:26 *** beatrootfarmer has joined #bitcoin-core-dev 490 2017-06-05T20:40:53 *** goatturneer has quit IRC 491 2017-06-05T20:41:15 *** goatturner has joined #bitcoin-core-dev 492 2017-06-05T20:44:23 *** beatrootfarmer has quit IRC 493 2017-06-05T20:44:43 *** laurentmt has joined #bitcoin-core-dev 494 2017-06-05T20:45:19 *** abpa has quit IRC 495 2017-06-05T20:47:05 <jtimon> cfields: thanks a lot! with your suggestion and another small change it seems the problem is gone 496 2017-06-05T20:47:10 <jtimon> re #10193 497 2017-06-05T20:47:11 <gribble> https://github.com/bitcoin/bitcoin/issues/10193 | scripted-diff: Remove #include foreach.hpp> by jtimon Â· Pull Request #10193 Â· bitcoin bitcoin Â· GitHub 498 2017-06-05T20:47:15 <jtimon> will update 499 2017-06-05T20:53:11 <jtimon> done 500 2017-06-05T21:01:11 *** abpa has joined #bitcoin-core-dev 501 2017-06-05T21:06:44 *** btcdrak has quit IRC 502 2017-06-05T21:14:05 *** cryptapus_afk has joined #bitcoin-core-dev 503 2017-06-05T21:29:05 *** chjj has quit IRC 504 2017-06-05T21:39:30 *** Chris_Stewart_5 has quit IRC 505 2017-06-05T21:43:12 *** chjj has joined #bitcoin-core-dev 506 2017-06-05T21:48:08 <cfields> jtimon: great :) 507 2017-06-05T21:49:21 <jtimon> :) 508 2017-06-05T21:50:04 <cfields> jtimon: err, you need to use the reverser thing everywhere, not just in tests 509 2017-06-05T21:50:29 <jtimon> maybe now it doesn't make sense for them to be separated, but maybe if I hadn't separated them I would still be stuck, so no regrets 510 2017-06-05T21:50:46 <jtimon> oh, I see 511 2017-06-05T21:52:20 <jtimon> I didn't run the python tests, but they should fail in that case, I'll wait for travis to fail 512 2017-06-05T21:52:42 <jtimon> then edit the scripted-diff commit to do the same thing everywhere 513 2017-06-05T21:53:11 <jtimon> although prevector tests will still need something different for the const reverse loop 514 2017-06-05T21:53:17 <jtimon> or I think so 515 2017-06-05T21:54:04 <jtimon> thanks for the heads up, I really thought it was just there because of how templates are used there 516 2017-06-05T21:55:05 <cfields> np. no, it's because of how a range-based-for binds to the expression 517 2017-06-05T21:55:06 <jtimon> then the new class sucks a little bit, perhaps I can do the same with auto and rbegin rend inlined 518 2017-06-05T21:55:23 <cfields> though I can't say that I 100% understand the semantics 519 2017-06-05T21:55:24 <jtimon> without needing a new class 520 2017-06-05T21:55:47 <jtimon> alright, let's wait for travis 521 2017-06-05T21:56:09 <jtimon> if it passes and you are still right maybe we can spot some missing tests 522 2017-06-05T21:56:47 <jtimon> I can run the python tests locally too, I'll do that 523 2017-06-05T22:05:01 <jtimon> jnewbery: what do you think about changing /tmp/bitcoin_test_runner_20170605_235927 to /tmp/bitcoin_test_runner/20170605_235927 ? 524 2017-06-05T22:06:01 <jtimon> not a big deal, just slightly easier to remove the whole folder, kudos for the clearer prefix 525 2017-06-05T22:06:11 <jnewbery> mildly against. Why add an extra directory layer? It just means I need to press the tab key one more time :) 526 2017-06-05T22:06:33 <jnewbery> why easier to remove the whole folder? 527 2017-06-05T22:06:39 <jtimon> fair enough, that was the question 528 2017-06-05T22:07:18 <jtimon> it wasn't unlikely that it was only more convenient for me 529 2017-06-05T22:07:37 <jtimon> the current prefix is good enough anyway 530 2017-06-05T22:09:53 <jtimon> cfields: I don't know about travis, but this passes python3 ./test/functional/test_runner.py -j56 --extended -x pruning,smartfees,maxuploadtarget 531 2017-06-05T22:10:07 <jnewbery> jtimon: cool. Thanks 532 2017-06-05T22:10:54 <cfields> jtimon: I think you need a new set of tests for reverse_iterator. I don't trust it with the simple for construct :( 533 2017-06-05T22:15:05 <jtimon> cfields: yep, more reason to simply rewrite the scripted-diff into something that uses rbegin/rend direclty instead of adding the new template 534 2017-06-05T22:15:20 <cfields> agreed 535 2017-06-05T22:15:45 <jtimon> but, still, if this shouldn't pass tests and it passes, perhaps we should open an issue asking for more tests 536 2017-06-05T22:17:01 <cfields> agree with that too :) 537 2017-06-05T22:17:44 <jtimon> I will try the version without the new class either way, it'll probably look smaller, just a more complex script to read if you use scripted-diff for review 538 2017-06-05T22:18:20 <jtimon> but the diff should look very simple as well 539 2017-06-05T22:19:25 <jtimon> cool 540 2017-06-05T22:24:37 *** trippysalmon has quit IRC 541 2017-06-05T22:26:49 *** Dyaheon has quit IRC 542 2017-06-05T22:29:35 *** Dyaheon has joined #bitcoin-core-dev 543 2017-06-05T22:32:53 *** trippysalmon has joined #bitcoin-core-dev 544 2017-06-05T22:52:26 *** talmai has quit IRC 545 2017-06-05T23:18:44 *** jnewbery has quit IRC 546 2017-06-05T23:20:03 *** goatturneer has joined #bitcoin-core-dev 547 2017-06-05T23:21:34 *** jnewbery_ has quit IRC 548 2017-06-05T23:23:38 *** goatturner has quit IRC 549 2017-06-05T23:26:05 <bitcoin-git> [bitcoin] MarcoFalke opened pull request #10535: [qa] fundrawtx: Fix shutdown race (master...Mf1706-qaFundrawRace) https://github.com/bitcoin/bitcoin/pull/10535 550 2017-06-05T23:29:44 <midnightmagic> wait was there an rc that I missed a gitian for..? 551 2017-06-05T23:29:53 <BlueMatt> 14.2? maybe? 552 2017-06-05T23:30:09 <midnightmagic> argh 553 2017-06-05T23:32:37 <achow101> midnightmagic: might as well not build it now since an rc2 is guaranteed and the rc1 codesigned bins won't be made 554 2017-06-05T23:32:56 <achow101> the version number wasn't bumped 555 2017-06-05T23:34:34 *** beatrootfarmer has joined #bitcoin-core-dev 556 2017-06-05T23:34:44 *** talmai has joined #bitcoin-core-dev 557 2017-06-05T23:35:10 <midnightmagic> achow101: I have the time, and I am one of those completionists you see sometimes who opens every crate in the game just because there might be a coupla gold in there.. 558 2017-06-05T23:35:46 <luke-jr> but there is no chance of gold in this case? :p 559 2017-06-05T23:36:01 <midnightmagic> The gold is figurative! 560 2017-06-05T23:38:05 *** goatturneer has quit IRC 561 2017-06-05T23:46:38 <luke-jr> midnightmagic: what does it represent in this case? 562 2017-06-05T23:56:26 <bitcoin-git> [bitcoin] practicalswift opened pull request #10536: Remove unreachable or otherwise redundant code (master...unreachable) https://github.com/bitcoin/bitcoin/pull/10536 563 2017-06-05T23:56:40 <midnightmagic> luke-jr: The reward of having opened every case. :) Also, I can say to people that I've built almost every gitian and therefore personally verified that almost every build is represented by source code as it is in the repo.