--- Day changed Mon Aug 13 2018 02:22 -!- danielhuman [~danielhum@221.239.69.250] has joined #secp256k1 02:22 -!- danielhuman [~danielhum@221.239.69.250] has quit [Remote host closed the connection] 02:24 -!- moondoggy29 [~moondoggy@128.199.117.132] has joined #secp256k1 02:26 -!- moondoggy29 [~moondoggy@128.199.117.132] has quit [Remote host closed the connection] 02:28 -!- jelly0 [~jelly@117.255.223.133] has joined #secp256k1 02:28 -!- jelly0 [~jelly@117.255.223.133] has quit [Remote host closed the connection] 02:37 -!- agris [~agris@gateway/tor-sasl/agris] has quit [Remote host closed the connection] 02:37 -!- agris [~agris@gateway/tor-sasl/agris] has joined #secp256k1 02:37 -!- insidious8 [~insidious@2001:e00:100:b45f:7af5:57ff:fe47:4f58] has joined #secp256k1 02:43 -!- insidious8 [~insidious@2001:e00:100:b45f:7af5:57ff:fe47:4f58] has quit [Ping timeout: 260 seconds] 04:38 -!- bumbar24 [~bumbar@37.237.59.247] has joined #secp256k1 04:40 -!- bumbar24 [~bumbar@37.237.59.247] has quit [Remote host closed the connection] 04:59 -!- Cajs14 [~Cajs@2.124.31.95] has joined #secp256k1 05:01 -!- Cajs14 [~Cajs@2.124.31.95] has quit [Remote host closed the connection] 05:55 -!- OPK6 [~OPK@103.209.178.216] has joined #secp256k1 05:57 -!- OPK6 [~OPK@103.209.178.216] has quit [Remote host closed the connection] 06:03 -!- duckgoose15 [~duckgoose@121.226.100.23] has joined #secp256k1 06:05 -!- duckgoose15 [~duckgoose@121.226.100.23] has quit [Remote host closed the connection] 08:21 -!- Levex [~Levex@58.48.50.226] has joined #secp256k1 08:26 -!- Levex [~Levex@58.48.50.226] has quit [Remote host closed the connection] 09:26 -!- deusexbeer [~deusexbee@095-129-172-082-dynamic-pool-adsl.wbt.ru] has joined #secp256k1 09:39 -!- \mSg9 [~\mSg@37.238.63.81] has joined #secp256k1 09:44 -!- \mSg9 [~\mSg@37.238.63.81] has quit [Remote host closed the connection] 10:19 -!- bluszcz28 [~bluszcz@27.22.48.92] has joined #secp256k1 10:23 -!- bluszcz28 [~bluszcz@27.22.48.92] has quit [Remote host closed the connection] 11:02 -!- plat_8 [~plat_@41.58.157.162] has joined #secp256k1 11:04 -!- plat_8 [~plat_@41.58.157.162] has quit [Remote host closed the connection] 12:17 -!- this [~this@27.156.142.111] has joined #secp256k1 12:18 -!- this [~this@27.156.142.111] has quit [Remote host closed the connection] 12:19 < roconnor> old: ecmult_32768g: min 8.68us / avg 9.09us / max 9.35us 12:19 < roconnor> new: ecmult_32768g: min 8.53us / avg 8.72us / max 8.85us 12:19 < andytoshi> whew 12:20 < roconnor> P.S. the failing checks in the PR seem transient to me. 12:20 < andytoshi> that's really good 12:20 -!- harding_ [quassel@2600:3c03::f03c:91ff:fe7b:78d1] has joined #secp256k1 12:20 < andytoshi> yeah, in rust-bitcoin we've got like a dozen things failed because of random network issues 12:20 < andytoshi> can you close and re-open the PR to re-trigger travis? 12:21 -!- cfields_ [~quassel@unaffiliated/cfields] has joined #secp256k1 12:21 < roconnor> done. 12:21 < andytoshi> lol, my text-on-libsecp256k1 script cost me 2.25 cents for that 12:22 < andytoshi> (3 texts - one for your comment, one for the close, one for the open) 12:22 < roconnor> :-( 12:22 < andytoshi> it's all good, i should really clean that up to be less text-happy ... or find a cheaper texting service 12:22 <@gmaxwell> 4% speedup is nice! 12:23 < roconnor> It's larger than I expected. 12:23 < roconnor> I haven't looked for the same speedup in ecmult_const 12:23 -!- jonasschnelli_ [~jonasschn@bitcoinsrv.jonasschnelli.ch] has joined #secp256k1 12:24 < roconnor> 'cause I only care about validation. :P 12:25 -!- Netsplit *.net <-> *.split quits: harding 12:28 < roconnor> andytoshi: I'm not even convinced that closing the PR has restarted travis. 12:28 < andytoshi> looks like it has 12:28 < andytoshi> (curious if other people think this is more or less obnoxious than force-pushing. on the one hand it looks weird and triggers github notifications; on the other hand it doesn't change the commit hash and force reviewers to sanity-check it) 12:28 <@gmaxwell> anyone with commit access to the repo can also trigger it to run again. 12:59 -!- Pasha [~Cory@unaffiliated/cory] has quit [Ping timeout: 240 seconds] 13:07 -!- Pasha [~Cory@unaffiliated/cory] has joined #secp256k1 13:32 -!- jonasschnelli_ [~jonasschn@bitcoinsrv.jonasschnelli.ch] has quit [Changing host] 13:32 -!- jonasschnelli_ [~jonasschn@unaffiliated/jonasschnelli] has joined #secp256k1 13:35 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 13:35 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #secp256k1 13:42 < roconnor> I'm not sure if this is important but @ https://github.com/bitcoin-core/secp256k1/blob/1e6f1f5ad5e7f1e3ef79313ec02023902bf8175c/src/tests.c#L230 the NULL passed to context_build will be dereferenced if the memory allocation fails. 13:43 <@gmaxwell> I don't think we care about null deref on malloc fail in the tests. 13:45 < roconnor> Technically it's undefined behavour meaning the tests could pass if malloc fails. ^_^ 13:53 < roconnor> Oh, secp256k1_ecmult_context_build is already called by secp256k1_context_create 13:53 < roconnor> so the line just doesn't anything. 13:54 <@gmaxwell> Read the comment above. 13:54 <@gmaxwell> It is intentionally redundant, because it's trying to see if you create redundantly if you'll leak memory. 13:54 <@gmaxwell> (which I believe is a bug that an earlier implementation had) 13:55 <@gmaxwell> actually detecting the leak is accomplished by running the tests inside valgrind, which is something we do. (the tests also have other valgrind specific instrumentation) 14:04 < roconnor> yep I think I get it now. 14:07 -!- tharkun22 [~tharkun@41.234.173.41] has joined #secp256k1 14:08 -!- Guest89575 [~LEI@ip-83-134-179-225.dsl.scarlet.be] has joined #secp256k1 14:11 -!- tharkun22 [~tharkun@41.234.173.41] has quit [Remote host closed the connection] 14:12 -!- Guest89575 [~LEI@ip-83-134-179-225.dsl.scarlet.be] has quit [Remote host closed the connection] 15:50 -!- agris_ [~agris@gateway/tor-sasl/agris] has joined #secp256k1 15:52 -!- agris [~agris@gateway/tor-sasl/agris] has quit [Ping timeout: 250 seconds] 15:54 -!- planetmaker10 [~planetmak@134.236.97.1] has joined #secp256k1 15:54 -!- planetmaker10 [~planetmak@134.236.97.1] has quit [K-Lined] 16:35 -!- labviking [~labviking@EAOcf-210p240.ppp15.odn.ne.jp] has joined #secp256k1 16:35 -!- prawnsalad24 [~prawnsala@140.224.80.156] has joined #secp256k1 16:36 -!- labviking [~labviking@EAOcf-210p240.ppp15.odn.ne.jp] has quit [Killed (Sigyn (Spam is off topic on freenode.))] 16:39 -!- prawnsalad24 [~prawnsala@140.224.80.156] has quit [Remote host closed the connection] 17:29 <@gmaxwell> Trivial optimization I randomly found on the internet, when we're reprojecting from gej to ge and using a ladder inverse, you can terminate the inverse computation early and save a field multiply: https://briansmith.org/ecc-inversion-addition-chains-01 17:48 -!- ken2812221_ [~ken281222@1.200.219.221] has joined #secp256k1 17:57 -!- Netsplit *.net <-> *.split quits: ken2812221 19:20 -!- Pasha [~Cory@unaffiliated/cory] has quit [Ping timeout: 260 seconds] 19:27 -!- Cory [~Cory@unaffiliated/cory] has joined #secp256k1 20:02 -!- sirnaysayer15 [~sirnaysay@27.130.156.90] has joined #secp256k1 20:05 -!- sirnaysayer15 [~sirnaysay@27.130.156.90] has quit [Remote host closed the connection] 20:19 -!- nickler [~nickler@185.12.46.130] has quit [Ping timeout: 276 seconds] 20:19 -!- Cory [~Cory@unaffiliated/cory] has quit [] 20:30 -!- nickler [~nickler@185.12.46.130] has joined #secp256k1 20:38 -!- ozymandias8 [~ozymandia@61.160.25.118] has joined #secp256k1 20:41 -!- nikow16 [~nikow@2402:800:6125:3057:6547:82a9:33b5:d699] has joined #secp256k1 20:42 -!- ozymandias8 [~ozymandia@61.160.25.118] has quit [Remote host closed the connection] 20:45 -!- nikow16 [~nikow@2402:800:6125:3057:6547:82a9:33b5:d699] has quit [Remote host closed the connection] 22:50 -!- ilbelkyr6 [~ilbelkyr@188.163.218.35] has joined #secp256k1 22:50 -!- ilbelkyr6 [~ilbelkyr@188.163.218.35] has quit [Killed (Unit193 (Spam is not permitted on freenode.))] 23:13 -!- agris__ [~agris@gateway/tor-sasl/agris] has joined #secp256k1 23:13 -!- agris_ [~agris@gateway/tor-sasl/agris] has quit [Remote host closed the connection] 23:26 -!- brethil [~brethil@159.192.245.230] has joined #secp256k1 23:26 -!- brethil [~brethil@159.192.245.230] has quit [K-Lined]