--- Log opened Mon Nov 25 00:00:18 2019 00:01 -!- jonatack_ [~jon@37.170.7.180] has joined #bitcoin-builds 00:04 -!- jonatack [~jon@37.164.37.175] has quit [Ping timeout: 265 seconds] 03:34 -!- MarcoFalke [~none@198.12.116.246] has quit [Quit: ZNC 1.7.1 - https://znc.in] 03:39 -!- MarcoFal- [~none@198.12.116.246] has joined #bitcoin-builds 03:44 -!- cfields [~cfields@unaffiliated/cfields] has quit [Remote host closed the connection] 03:48 -!- cfields_ [~cfields@unaffiliated/cfields] has joined #bitcoin-builds 05:04 < wumpus> elichai2: you found a way to do big-endian travis runs? that's exciting 05:05 < elichai2> yes 05:05 < elichai2> wumpus: https://travis-ci.org/elichai/bitcoin/jobs/616656410#L36 :) 05:05 < elichai2> I'm opening a PR now :) 05:09 < wumpus> so they have actual s390x hardware? isn't that, like, ancient mainframe stuff? 05:10 < elichai2> yeah they have some partenrship with IBM and Canonical 05:10 < elichai2> "The initial infrastructural build capacity has a total of 40 concurrent jobs on IBM Z (s390x) and 64 concurrent jobs on IBM Power (ppc64le) split between travis-ci.org and travis-ci.com." 05:10 < wumpus> wonder if it's powerful enough to compile and run bitcoin at all, but we'll see :) 05:11 < elichai2> wumpus: already did in my travis 05:11 < elichai2> https://travis-ci.org/elichai/bitcoin/builds/616656404 05:12 < wumpus> ohh! I thought you just tested that it was really big-endian 05:12 < wumpus> the unit tests even pass as-is 05:13 < elichai2> no no, I added `lscpu` so we can keep seeing that. I tested that it's actually big endian in rust first actually heh https://travis-ci.org/elichai/Test-travis-BE/jobs/616645985 05:14 < wumpus> also "Checking whether byte ordering is bigendian... (cached) yes" 05:14 < elichai2> altough cargo crashed on travis, so I had to cross compile localy and push a binary 05:14 < elichai2> oh haven't noticed that. cool :) 05:15 -!- davterra [~dulyNoded@91.132.136.84] has joined #bitcoin-builds 05:15 < elichai2> pretty exciting hehe 05:19 < wumpus> especially for serialization changes this is very good to have 05:47 < elichai2> should I try to enable depends too? 06:02 < wumpus> dunnoooo maybe later, depends on how long that takes, but you already have a ACK by MarcoFalke so maybe not change too much 06:04 < wumpus> definitely later, i don't think depends even supports s390x right now 06:08 < elichai2> oh heh, he 06:08 < elichai2> *he's the one that asked for changes too :) 06:08 < elichai2> I guess i'll try to add depends+functional and simplify some stuff late 06:09 < wumpus> you might be the first person ever to run bitcoin on s390x 06:09 < elichai2> if so i'm quite amazed it just worked 06:10 < wumpus> me too, though, i did testing on mips_be on qemu a long time back, but still surprised it just works 08:01 < elichai2> of course that you can't get `g++-s390x-linux-gnu` inside of `s390x` g++-s390x-linux-gnu 08:02 < elichai2> https://packages.ubuntu.com/bionic/g++-s390x-linux-gnu 08:02 * elichai2 facepalm 08:20 < elichai2> is there a built-in way to force `depends` to use `clang` instead of `gcc`? 08:31 < elichai2> I just can't seem to make boost use clang :( https://travis-ci.org/elichai/bitcoin/jobs/616766260#L844 08:32 < elichai2> how can I make all the depends stuff verbose? 08:32 < dongcarl> `V=1` 08:33 < dongcarl> elichai2: looking at your thing now 08:33 < dongcarl> did you do `HOST=s390x-linux-gnu`? 08:33 < elichai2> that's it? can I export that? 08:33 < elichai2> yes 08:33 < elichai2> that's the host 08:34 < dongcarl> Hmmm something is inserting an `-unknown-` in there 08:34 < dongcarl> I'm guessing the g++ binary is called `s390x-linux-gnu-g++`? 08:34 < elichai2> i'm not sure the unknown is the problem. that binary doesn't exist AFAICT 08:34 < elichai2> look here: https://packages.ubuntu.com/bionic/g++-s390x-linux-gnu it's only available for amd64/i386 08:34 < elichai2> not for s390x 08:35 < dongcarl> OH I SEE 08:35 < dongcarl> One sec 08:37 < dongcarl> elichai2: what happens when you don't set `HOST`? 08:37 < elichai2> i'll try 08:39 < fanquake> dongcarl What's the status of #16519 ? 08:40 < fanquake> Still waiting for a new release I assume. 08:40 < fanquake> Any idea when that might be 08:41 < dongcarl> fanquake: Not an improvement for user experience until new release, but we can do the user-experience-improving stuff after release, and just merge the practical parts here... 08:41 < dongcarl> release = guix release not bitcoin release 08:42 < dongcarl> I'm gunna work on tending to my chain of guix PRs this week 08:42 < fanquake> Cool. 08:42 < fanquake> I think #16392 is at the point where it can be scrutinized. 08:43 < fanquake> Could pull f2368e01fe27d03c7666bb5d9fd45c7aafd3378e out on its own if wanted. 08:45 < dongcarl> looking 08:47 < elichai2> dongcarl: seems to help for that problem heh, now I need to revert all my changes back to gcc/g++, will not providing a `HOST` hurt the ability to also use the QEMU option in the ci script? 08:48 < dongcarl> elichai2: can you remind me what the QEMU option in the ci script does? 08:48 < elichai2> theoretically it lets you run the ci script locally, and it will use qemu if your not travis (i.e. setting some env var) 08:49 < elichai2> unless I'm not understanding it correctly MarcoFal-: 08:49 < dongcarl> elichai2: that should work fine 08:50 < elichai2> so why do we even have that HOST var? 08:50 < dongcarl> elichai2: for cross compiling 08:50 < dongcarl> elichai2: you're not cross compiling 08:50 < elichai2> what *does* cross compile? (i.e. the CI never cross compiles, right?) 08:51 < dongcarl> elichai2: gitian and guix builds 08:51 < elichai2> :) Thanks 08:51 < dongcarl> np 08:51 < dongcarl> Also, there are just crazy people who cross-compiles 08:51 < dongcarl> Who knows 08:52 < elichai2> actually in rust it's damn too easy with https://github.com/rust-embedded/cross 08:52 < dongcarl> yeah yeah 08:54 < elichai2> well apparently boost just got built in s390x without a problem (well after your help heh) let's see if all the depe3nds will just work. and while I'm writing it the qt had failed lol 08:54 < dongcarl> Yeah... not sure if that's a supported platform for qt? 08:57 < elichai2> it works when installed as a package, i'll start investigating the qt code to see if it's somehow jut a missing precompiled macro 08:58 < dongcarl> elichai2: yeah if you can't work it out show me the exact error 08:58 < elichai2> thanks, i'll give it a try first 09:02 < elichai2> seems like it can't find `zconf.h` https://travis-ci.org/elichai/bitcoin/jobs/616777596#L4727 but i'll leave it for tomorrow :) 09:41 -!- jonatack_ [~jon@37.170.7.180] has quit [Quit: jonatack_] 09:42 -!- jonatack [~jon@37.170.7.180] has joined #bitcoin-builds 09:54 < fanquake> elichai2 interesting. Looks like qt finds zlib during it's configure phase at leasy 09:54 < fanquake> *least 10:16 < MarcoFal-> [11:48] elichai2: can you remind me what the QEMU option in the ci script does? 10:16 < MarcoFal-> Yes, it allows me to run the ci system on my aarch64 "farm" 10:17 < MarcoFal-> I'll probably merge now and then figure it out from there 10:17 < MarcoFal-> You might also have to add cross compile packages 10:19 < MarcoFal-> oh wait, could you remove the lscpu please, elichai2 10:19 < MarcoFal-> (or move it to where we do `free`) 10:22 < MarcoFal-> btw, can anyone see my messages? 10:22 < fanquake> MarcoFal- yea 10:22 < MarcoFal-> kool 10:40 -!- MarcoFal- [~none@198.12.116.246] has quit [Remote host closed the connection] 10:40 -!- MarcoFalk [~none@198.12.116.246] has joined #bitcoin-builds 10:41 -!- MarcoFalk [~none@198.12.116.246] has quit [Remote host closed the connection] 10:45 -!- MarcoFalke [~none@198.12.116.246] has joined #bitcoin-builds 11:56 < MarcoFalke> Anyone want to scroll through https://github.com/bitcoin/bitcoin/pull/17549 and give it an ACK? 11:57 < MarcoFalke> it removes a linter, so the builds on travis will be faster. Should be an easy ACK 12:02 -!- jonatack_ [~jon@37.173.132.239] has joined #bitcoin-builds 12:03 < elichai2> MarcoFalke: I'll try to do it together with no depends(if I'll be able to make them work) 12:03 < MarcoFalke> Last time I tried depends on s390x they didn't work 12:03 < MarcoFalke> good luck, though 12:04 < fanquake> ? 12:05 -!- jonatack [~jon@37.170.7.180] has quit [Ping timeout: 276 seconds] 12:08 < MarcoFalke> at least we removed openssl, which was a pain to cross compile with our depends 12:20 < MarcoFalke> Ah you need to run trough full depends if you want to cross compile 13:25 < hebasto> How does Qt uic work in the makefile rule (line 292 of Makefile.qt.include); I cannot see `uic' invocation at all... 13:26 < hebasto> ? 13:32 < fanquake> hebasto scroll down to 355 13:36 < hebasto> fanquake: thanks! 13:45 < hebasto> fanquake: it seems comment in line 291 should be "forms/foo.ui -> forms/ui_foo.h", no? 13:46 -!- _jerry [~jerry@142.134.28.200] has joined #bitcoin-builds 13:46 -!- bitbanger [~jerry@142.134.28.200] has joined #bitcoin-builds 13:47 -!- bitbanger [~jerry@142.134.28.200] has quit [Client Quit] 15:22 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has quit [Ping timeout: 260 seconds] 15:46 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has joined #bitcoin-builds 19:57 -!- jonatack_ [~jon@37.173.132.239] has quit [Read error: Connection reset by peer] 20:49 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 246 seconds] 21:22 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #bitcoin-builds 21:45 -!- _jerry [~jerry@142.134.28.200] has quit [Ping timeout: 265 seconds] 22:28 -!- jonatack [~jon@37.173.132.239] has joined #bitcoin-builds --- Log closed Tue Nov 26 00:00:19 2019