--- Log opened Fri Feb 12 00:00:27 2021
00:22 -!- Kiminuo [~Kiminuo@141.98.103.228] has joined #rust-bitcoin
01:39 -!- belcher_ is now known as belcher
02:08 < sgeisler> the powerset of our features is kinda huge :/ when I'm testing stuff fully I now run the test suite in 256 configurations (64x1.29, 64xstable, 128xnightly) that takes a bit more than an hour (not optimally parallelized to avoid compiling the same deps too often), but assuming you wanted to do this for each commit and not just `pr/xyz/merge` that becomes unpractical rather quickly.
02:09 < sgeisler> andytoshi do you know if secp-lowmemory should ever have any influence on our tests in rust-bitcoin? Does it change the API of secp?
02:19 < sgeisler> after not really getting andytoshi's git-scripts to run I built some tooling of my own, not focused on git  (that's done by a bash script for now), just to orchestrate the tests. If anyone is interested: https://github.com/sgeisler/testinator
03:19 -!- Laverne35Rolfson [~Laverne35@static.57.1.216.95.clients.your-server.de] has joined #rust-bitcoin
03:30 -!- midnight [~midnight@unaffiliated/midnightmagic] has quit [Ping timeout: 240 seconds]
03:33 -!- midnight [~midnight@unaffiliated/midnightmagic] has joined #rust-bitcoin
04:12 -!- tibo [~tibo@2400:4050:2a83:7000:6990:d3ed:dacb:6518] has quit []
05:40 < andytoshi> sgeisler: no, it should have no effect whatsoever
05:40 < andytoshi> wow @ 256 combinations!
05:40 < sgeisler> nice, that should cut the time in half ^^
05:41 < andytoshi> i have a 64-core machine that I use for this..although cargo is really really bad at being run in parallel
05:41 < andytoshi> but tbh I just run no feature, then every individual feature, then all features
05:41 < andytoshi> and run the fuzztests with a large number of iterations
05:42 < andytoshi> and i only run 1.29 on the tip because certain people here never test their commits with 1.29 and i got tired of chasing it down (because realistically when would we ever be bisecting with 1.29 anyway..)
05:42 < sgeisler> (hides)
05:42 < andytoshi> hehehe
05:43 < sgeisler> do you fuzz on multiple rust versions?
05:43 < andytoshi> IIRC i would take a bit under 10 minutes per commit to run my tests
05:43 < andytoshi> no, i only fuzz on nightly
05:43 < andytoshi> i didn't even know that i could fuzz on other versions
05:43 < sgeisler> ok, same here, idk if it works on others, just wanted to check
05:43 < andytoshi> i suspect that it doesn't work... and if it does, it probably wouldn't accomplish anything. just be slower.
05:44 < sgeisler> I mean ideally at least, as long as rust is sound ^^
05:44 < andytoshi> yep
05:44 < sgeisler> especially since we don't use unsafe
05:44 < andytoshi> right
05:45 < andytoshi> i've been working on replacing my build system with one that's waay more paralell
05:45 < andytoshi> but cargo gets deadlocked a lot and i've gotten frustrated trying to debug it
05:45 < sgeisler> well, I just copy the repo as often as I want parallel runs, does it still lock?
05:45 < andytoshi> yes
05:45 < sgeisler> ah, for the global inde
05:45 < sgeisler> x
05:46 < andytoshi> yeah. it's infuriating
05:46 < andytoshi> it's also ungodly slow because it rebuilds all the deps for every clean checkout
05:46 < andytoshi> why does it have a freeking global lock if it's not even going to share work? i dunno
05:47 < andytoshi> i'm considering moving all the projects to bazel instead of cargo, but i haven't gotten around to scoping that out
05:48 < sgeisler> wouldn't that cause complete incompatibility with the ecosystem or can bazel just work with cargo but is more intelligent?
05:49 < andytoshi> good question, i would hope that cargo is not so incompatible with everything else
05:49 < andytoshi> but historically it has been, even different versions of cargo don't work with each other
05:50 < andytoshi> in any case .. as long as 'cargo publish' works i expect that's all we need to be "compatible" with the rest of the ecosystem
05:51 < sgeisler> did you try this?: https://doc.bccnsoft.com/docs/rust-1.36.0-docs-html/cargo/reference/unstable.html#no-index-update
05:51 < andytoshi> oo no, that's a good idea
05:52 < andytoshi> it's encouraging that they say crater uses this, becuase i'm sure crater runs into the same bugs i do
05:52 < sgeisler> might not work for 1.29 since it's unstable in 1.36, but for nightly and stable it might work
06:02 < andytoshi> i can live with that
06:35 -!- Laverne35Rolfson [~Laverne35@static.57.1.216.95.clients.your-server.de] has quit [Remote host closed the connection]
06:38 -!- Brandi33Nolan [~Brandi33N@static.57.1.216.95.clients.your-server.de] has joined #rust-bitcoin
06:42 -!- Kiminuo [~Kiminuo@141.98.103.228] has quit [Quit: Leaving]
06:45 -!- jonatack [~jon@37.169.23.248] has joined #rust-bitcoin
06:54 < BlueMatt> andytoshi: to be fair, I only intend to use it in the C bindings and especially in WASM where, like, i dont trust rand
10:39 -!- jonatack [~jon@37.169.23.248] has quit [Ping timeout: 256 seconds]
10:41 -!- jonatack [jon@gateway/vpn/airvpn/jonatack] has joined #rust-bitcoin
10:51 -!- jonatack [jon@gateway/vpn/airvpn/jonatack] has quit [Ping timeout: 240 seconds]
10:53 -!- jonatack [~jon@37.169.23.248] has joined #rust-bitcoin
12:38 < BlueMatt> lol wtf why does rust-bitcoin have no problem using test::Bencher but my own crate has an unresolved import
12:41 < BlueMatt> oh, lol missing feature flag for it
12:41 -!- Brandi33Nolan [~Brandi33N@static.57.1.216.95.clients.your-server.de] has quit [Ping timeout: 256 seconds]
13:19 -!- jonatack [~jon@37.169.23.248] has quit [Read error: Connection reset by peer]
13:21 -!- jonatack [~jon@37.169.23.248] has joined #rust-bitcoin
15:50 -!- jonatack_ [~jon@37.166.60.165] has joined #rust-bitcoin
15:53 -!- jonatack [~jon@37.169.23.248] has quit [Ping timeout: 240 seconds]
17:58 -!- belcher_ [~belcher@unaffiliated/belcher] has joined #rust-bitcoin
18:02 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 272 seconds]
19:36 -!- jonatack_ [~jon@37.166.60.165] has quit [Ping timeout: 246 seconds]
--- Log closed Sat Feb 13 00:00:27 2021