--- Log opened Thu Jul 07 00:00:13 2022
00:41 -!- stickies-v [sid544753@id-544753.uxbridge.irccloud.com] has quit [Ping timeout: 244 seconds]
00:42 -!- berryyoghurt [sid553661@id-553661.helmsley.irccloud.com] has quit [Read error: Connection reset by peer]
00:42 -!- berryyoghurt [sid553661@id-553661.helmsley.irccloud.com] has joined #secp256k1
00:42 -!- stickies-v [sid544753@id-544753.uxbridge.irccloud.com] has joined #secp256k1
06:43 < andytoshi> sipa: kudos for your politeness to the haskell guy!
08:06 -!- halosghost [~halosghos@user/halosghost] has joined #secp256k1
08:16 < real_or_random> yeah indeed, your reply was more elaborate than mine, at least initally ^^
08:18 < real_or_random> sipa: wanna review https://github.com/bitcoin-core/secp256k1/pull/1120? it's not yet the PR about API docs I mentioned (still working on this) but would be still nice to have for the release
08:55 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has quit [Remote host closed the connection]
08:56 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has joined #secp256k1
09:00 -!- sigurdson[m] [~sigurdson@2001:470:69fc:105::2:25f5] has quit [Quit: You have been kicked for being idle]
11:14 < real_or_random> hm that code is strange... we run the tests with a randomized context in 1/4 of runs Oo https://github.com/bitcoin-core/secp256k1/blob/master/src/tests.c#L7090-L7093
11:15 < sipa> heh
11:20 < real_or_random> I mean.. anyway it's somewhat strange to flip a coin here but two coins?! ^^
11:20 < sipa> Yeah, this looks like it's code that got duplicated somehow.
11:20 < real_or_random> I checked, it's there since 2015
11:20 < sipa> I bet that git blame will reveal I added it.
11:21 < real_or_random> not sure what would better... running the ecmult tests twice sounds wrong too... maybe randomize in 15/16 cases would be a quick improvement?
11:21 < sipa> i'm fine with either
11:22 < sipa> if we want to make sure the code works with and without randomization, we could just run all ecmult tests twice
11:22 < sipa> i think we should also do a general "recalibration" of how many times every test gets run at some point
11:23 < real_or_random> yeah though how do you do this? profiler?
11:23 < sipa> or just inserting printf with accurate timestamps
11:24 < real_or_random> yeah, ok, I'd call this a profiler :P
11:24 < sipa> fair
12:06 < real_or_random> sipa: by the way, I discovered some kind of code smell when compiling manually
12:06 < real_or_random> so we usually pass -Wall -Wno-unused-function ... see what happens if you enable -Wunused-function
12:07 < real_or_random> it won't only report unused functions but in the precomputed .c files, it will also report many declared but undefined functions 
12:08 < real_or_random> basically when we #include group.h and the like we get a bunch of function declarations hat have no corresponding definition because we never include the group_impl.h 
12:09 < real_or_random> this has been overlooked because the warnings are surpressed by  -Wunused-function
12:09 < real_or_random> I don't even know if we should do anything about it. It's valid C and anyway have  -Wno-unused-function . But I was surprised to see this.
12:10 -!- jeremyrubin [~jeremyrub@ec2-44-199-24-18.compute-1.amazonaws.com] has joined #secp256k1
12:10 -!- instagibbs [~instagibb@pool-100-15-132-66.washdc.fios.verizon.net] has joined #secp256k1
12:10 -!- jarolrod [sid475272@id-475272.uxbridge.irccloud.com] has joined #secp256k1
12:10 -!- gleb1 [~gleb@178.150.137.228] has joined #secp256k1
12:10 -!- achow101 [~achow101@user/achow101] has joined #secp256k1
12:10 -!- dunxen [~dunxen@2001:470:69fc:105::1:fec1] has joined #secp256k1
12:12 -!- jonatack [~jonatack@user/jonatack] has joined #secp256k1
12:12 < sipa> (I just mentioned the secp256k1 IRC meetings in the Bitcoin Core IRC meeting, if anyone is surprised by the influx here...)
12:14 < hebasto> maybe put meeting time into this channel topic?
12:14 < real_or_random> yeah, I also wanted to add it to https://bitcoincore.org/en/meetings/  
12:34 < sipa> real_or_random: I'd say we shouldn't do anything about unused functions, except maybe from time to time go inspect if there aren't any actually-unconditionally-unused functions which have become pointless.
12:35 < sipa> Most of them I think are just due to use in optional modules, or being test-only.
12:45 < real_or_random> sipa: fully agree but the warnings I'm talking about were not warnings about unused functions
12:46 < sipa> oh
12:46 < real_or_random> they were warnings about functions which are declared and never used (which happen to be covered by -Wunused-function)
12:47 < real_or_random> if we include just the .h files but not the _impl.h files in the precomputed_.c we get the declarations but not the definitions
12:52 < sipa> hmm, that sounds pretty harmless, but i guess it would be cleaner not to do that
12:54 < real_or_random> this matches my thinking. right now I wouldn't care.
12:54 < real_or_random> if we had caught it during review, yeah, but now... ^^
15:10 -!- halosghost [~halosghos@user/halosghost] has quit [Quit: WeeChat 3.5]
21:11 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has quit [Remote host closed the connection]
21:12 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has joined #secp256k1
22:10 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has quit [Remote host closed the connection]
22:10 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has joined #secp256k1
--- Log closed Fri Jul 08 00:00:14 2022