#archlinux32 | Logs for 2020-04-28
Back
[00:20:23] -!- KeiraT has quit [Ping timeout: 240 seconds]
[00:30:46] -!- KeiraT has joined #archlinux32
[01:13:03] -!- KeiraT has quit [Ping timeout: 240 seconds]
[01:16:02] -!- KeiraT has joined #archlinux32
[03:44:34] -!- rcf has quit [Quit: WeeChat 2.7]
[03:45:18] -!- rcf has joined #archlinux32
[03:53:29] -!- petris has joined #archlinux32
[03:53:50] -!- petris has quit [Client Quit]
[04:03:04] -!- petris has joined #archlinux32
[04:08:30] -!- torv has quit [Remote host closed the connection]
[04:08:55] -!- torv has joined #archlinux32
[08:19:27] -!- torv has quit [Remote host closed the connection]
[08:22:11] -!- torv has joined #archlinux32
[09:12:14] -!- torv has quit [Remote host closed the connection]
[09:12:44] -!- torv has joined #archlinux32
[09:38:14] -!- bill-auger has quit [Ping timeout: 240 seconds]
[09:38:56] -!- bill-auger has joined #archlinux32
[16:04:01] -!- abaumann has joined #archlinux32
[16:04:01] <buildmaster> Hi abaumann!
[16:04:01] <buildmaster> !rq abaumann
[16:04:02] <phrik> buildmaster: <abaumann> it's modern software in the end.. it _has_ to compile that long in order to get good.. ;-)
[16:04:05] -!- deep42thought has joined #archlinux32
[16:04:05] <buildmaster> Hi deep42thought!
[16:04:05] <buildmaster> !rq deep42thought
[16:04:06] <phrik> buildmaster: <deep42thought> computers ranked by dirt: smokers, car mechanics, bedroom, ...
[16:04:10] <abaumann> hi deep42thought
[16:04:12] <deep42thought> Hi abaumann!
[16:04:20] <abaumann> I have a hanging sanity-check on the buildmaster..
[16:04:32] <abaumann> since two hours or so, at least
[16:04:32] <deep42thought> dang
[16:04:33] <deep42thought> my fault
[16:04:58] <abaumann> but not a touched file? I thought I checked for that.
[16:05:03] <deep42thought> since the reboot of the master mirror
[16:05:10] <abaumann> ah.
[16:05:12] <deep42thought> builder/work/intentions were there
[16:05:21] <deep42thought> should work again, now
[16:05:26] <abaumann> ok
[16:05:27] <abaumann> thanks.
[16:05:29] <abaumann> :-)\
[16:05:33] <abaumann> yep.
[16:05:35] <abaumann> it does
[16:05:55] <deep42thought> should I add some status to *why* the jobs hang?
[16:06:14] <abaumann> now we now.. till we forget again ;-)
[16:06:22] <abaumann> *know
[16:11:18] <buildmaster> pentium4/vtk is broken (says eurobuild6-6): https://archlinux32.org
[16:12:46] <buildmaster> i486/gdal is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:14:51] <buildmaster> pentium4/netcdf-fortran is broken (says eurobuild6-6): https://archlinux32.org
[16:14:52] <buildmaster> i486/netcdf-fortran is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:17:29] <buildmaster> i486/intel-opencl-clang is broken (says nlopc46-i486bs1): https://archlinux32.org
[16:18:40] <deep42thought> I just read a lengthy note on why "till" is wrong
[16:20:24] <buildmaster> i686/netcdf-fortran-openmpi is broken (says eurobuild6-5): https://archlinux32.org
[16:25:06] <abaumann> I'm using till until proven wrong.. ;-)
[16:25:14] <deep42thought> !grab abaumann
[16:25:14] <phrik> deep42thought: Bazinga!
[16:29:49] <buildmaster> i486/netcdf-fortran-openmpi is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:33:57] <buildmaster> i686/vtk is broken (says rechenknecht): https://archlinux32.org
[16:53:05] <buildmaster> i486/rime-cantonese is broken (says nlopc46-i486bs1): https://archlinux32.org
[16:59:52] <buildmaster> i486/bash-language-server is broken (says nlopc46-i486bs1): https://archlinux32.org
[17:06:35] <buildmaster> i486/pymol is broken (says nlopc46-i486bs1) - already flagged out-of-date upstream on 2019-05-27: https://archlinux32.org
[17:12:23] <buildmaster> i486/opencsg is broken (says nlopc46-i486bs1): https://archlinux32.org
[17:17:15] <buildmaster> i486/mesa-demos are broken (says nlopc46-i486bs1): https://archlinux32.org
[17:23:14] <buildmaster> i486/frogatto is broken (says nlopc46-i486bs1): https://archlinux32.org
[17:33:06] <buildmaster> i486/qbs are broken (says nlopc46-i486bs1): https://archlinux32.org
[17:37:48] * abaumann considers to add a herd of i486 slaves just to go through all the failures faster..
[18:00:19] <deep42thought> hehe
[18:01:56] <deep42thought> we should do some blacklist-rounds, again, I guess
[18:26:48] <buildmaster> i686/svt-av1 is broken (says rechenknecht): https://archlinux32.org
[18:28:10] <buildmaster> pentium4/svt-av1 is broken (says eurobuild6-6): https://archlinux32.org
[18:29:58] <buildmaster> i686/imagemagick6 is broken (says nlopc46): https://archlinux32.org
[18:30:30] <buildmaster> pentium4/imagemagick6 is broken (says nlopc46): https://archlinux32.org
[19:02:06] <abaumann> can we empty the i486 build queue?
[19:02:23] <abaumann> I fear, blacklisting doesn't help, it just keeps on building and building old stuff.
[19:16:58] -!- isacdaavid has joined #archlinux32
[19:22:14] <deep42thought> you can manually remove stuff from the mysql database
[19:22:26] <deep42thought> but I *think*, blacklisting will remove it from the build-list, too
[19:22:32] <deep42thought> at least, it should
[19:24:19] <abaumann> I would prefer a priority queue, so when I priotize something, the it should be build as next package
[19:24:30] <abaumann> otherwise I can not bootstrap some packages..
[19:24:42] <deep42thought> this is, how it works
[19:24:48] <deep42thought> but it still respects dependencies
[19:24:55] <deep42thought> this is probably, what causes you trouble
[19:25:03] <abaumann> so, ijs depends on something?
[19:25:09] <abaumann> that's why it doesn't get scheduled.
[19:25:36] <deep42thought> ijs is not on the build list
[19:25:52] <abaumann> I scheduled it half a dozen times since yesterday. :-)
[19:26:05] <abaumann> it gets promptly built on i686 and pentium4, but never on i486
[19:26:16] <deep42thought> sounds like blacklist
[19:26:22] <abaumann> it _has_ a circular dependency on ghostscript too
[19:26:30] <deep42thought> blacklist/i486/rust/ijs
[19:26:32] <deep42thought> there you go
[19:26:34] <abaumann> argh.
[19:26:53] <deep42thought> "blocked by ghostscript"
[19:27:02] <abaumann> so, why is in in rust?
[19:27:14] <deep42thought> because ghostscript was blocked due to rus?
[19:27:16] <deep42thought> *rust
[19:27:32] <abaumann> I doubt they were so crazy to make it depend on rust! :->
[19:27:40] <deep42thought> but on librsvg, maybe?
[19:27:41] <abaumann> I suspect some rust bindings
[19:28:16] <deep42thought> ghostscript: "blocked by gtk3"
[19:28:20] <abaumann> yeah, maybe. but just if you want rvg images in ghostscript. this is another cycle.
[19:28:27] <abaumann> there are cycles everywhere. *grr*
[19:28:54] <deep42thought> check the content of the blacklist files - I usually left a comment there, why the respective package is blacklisted
[19:29:28] <abaumann> ok.
[19:29:33] <abaumann> completely forgot about those.
[19:29:38] <deep42thought> :-)
[19:34:15] <abaumann> all good :-)
[19:51:27] <deep42thought> you forgot to remove ghostscript from the blacklist ;-)
[19:53:16] <abaumann> oh. no. :-)
[19:53:23] <abaumann> I need a helper-bot ;-)
[19:53:33] <abaumann> stopping me from doing silly things..
[19:53:44] <abaumann> ..hang on - I have a helper-bot :-)
[19:54:13] <abaumann> yeah. my i486 slave builds cmake now..
[19:54:26] <abaumann> ..ok, that's something I also need soon..
[19:55:02] * abaumann thinks watching Youtube documentations and coding don't work well together
[20:03:59] <deep42thought> there is not much, a helper bot could do: it's absolutely "legal" to have modifications for a package and simultanously blacklist the package for certain architectures
[20:04:41] <abaumann> I was thinking more of an 'Did you think about the blacklist?' as pre-git-commit-hook. ;-)
[20:04:54] <deep42thought> this is a two-liner :-D
[20:05:48] <abaumann> "You do not build anything currently - abort whatever you are doing"
[20:05:52] <abaumann> I'm lucky :-)
[20:06:07] <deep42thought> :-D
[20:06:18] <deep42thought> abort, abort, abort
[20:06:20] <abaumann> i686/ghostscript on eurobuild6-6 and eurobuild6-2
[20:06:26] <abaumann> pentium4/ghostscript on eurobuild6-1
[20:06:35] <abaumann> and an error on the i486 build slave.
[20:06:40] <abaumann> this doesn't sound right.
[20:07:06] <abaumann> at least I can test my PKGBUILD-sed-fu :-)
[20:08:35] <buildmaster> i486/texmacs are broken (says nlopc46-i486bs1): https://archlinux32.org
[20:27:11] <buildmaster> any/golang-gopkg-yaml.v2 is broken (says eurobuild6-6): https://archlinux32.org
[20:59:51] <deep42thought> abaumann: have you successfully built ghostscript on i486?
[21:00:45] <deep42thought> ah, yes, I see it :-)
[21:01:54] <abaumann> yep :-)
[21:02:02] <abaumann> and we cleaned the blacklist: in parallel :-)
[21:02:08] <deep42thought> oh, we did?
[21:02:09] <deep42thought> sry
[21:02:12] <abaumann> easy :-)
[21:02:13] <deep42thought> I was first :-)
[21:02:18] <abaumann> deletes are idempotent.
[21:02:22] <abaumann> in git
[21:02:33] <deep42thought> delete^2=delete
[21:02:49] <deep42thought> but beware, delete != 1
[21:02:49] <abaumann> would be funny if deleting twice would make things reappear. ;-)
[21:03:36] <deep42thought> deleting once deletes the file, deleting twice formats the partition, deleting three times formats the whole drive and deleting four times shreds the disk physically
[21:03:48] <abaumann> !grab deep42thought
[21:03:48] <phrik> abaumann: Bingpot!
[21:04:24] <abaumann> oh, no. Not firefox, please, please not firefox
[21:04:31] * abaumann kills a firefox build on the build slave..
[21:10:42] <buildmaster> pentium4/firefox is broken (says eurobuild6-1): https://archlinux32.org
[21:14:26] <deep42thought> abaumann: there are nicer ways to abort such a build ;-)
[21:14:35] <abaumann> reboot? ;-)
[21:16:08] <buildmaster> i486/ginac is broken (says nlopc46-i486bs0): https://archlinux32.org
[21:16:17] <deep42thought> return-assignment ABORT
[21:16:25] <deep42thought> and *then* you can kill the build
[21:16:45] <abaumann> ah.
[21:16:47] <abaumann> ok.
[21:16:51] <abaumann> WARNING: Project specifies a minimum meson_version '>= 0.41.0' but uses features which were added in newer versions:
[21:16:54] <abaumann> libxkbcommon
[21:17:10] <abaumann> if the Xorg group continues on this path, we will not have X in the future!
[21:17:11] <deep42thought> no problem, it merely prevent creating that false-positive build error
[21:18:02] <abaumann> You can disable X11 support with -Denable-x11=false.
[21:19:37] <abaumann> -Ddisable_enabled_not_X11=not_false
[21:19:57] <deep42thought> lol
[21:20:13] <abaumann> people doing double negation fail job interviews. :->
[21:20:25] <abaumann> they stress my brain.. :-)
[21:22:11] <abaumann> ah, that's the problem with the blacklist, you have to hunt down all dependencies when un-blacklisting.. but well..
[21:22:15] <abaumann> forgot diffoscope. :-)
[21:22:24] * abaumann does a git pull
[21:22:26] <deep42thought> no issue, find+grep help a lot
[21:22:52] <deep42thought> but this way, we easily see, what is blacklisted and why - no complicated dependency tracking
[21:23:08] <abaumann> yeah. exactly
[21:23:20] <abaumann> not to mention the performance improvement in the database.
[21:26:56] <abaumann> ../libxkbcommon/src/scanner-utils.h:62:13: warning: format â%ldâ expects argument of type âlong intâ, but argument 8 has type âint64_tâ {aka âlong long intâ} [-Wformat=] 62 | "%s:%zu:%zu: " fmt "\n", \
[21:27:01] <abaumann> yep. the code is a disaster
[21:31:04] <buildmaster> i486/libxkbcommon is broken (says eurobuild6-7-i486): https://archlinux32.org
[21:31:39] <buildmaster> i686/chromium is broken (says nlopc46): https://archlinux32.org
[21:31:56] <abaumann> it's like the code in nfs-utils: everytime you post some basic C-coding mistakes the next release has new ones :->
[21:32:40] <deep42thought> first law of coding: the conservation of number of stupid mistakes
[21:33:27] <abaumann> and: actually compile the code on something else than latest ubuntu before doing a 'git commit'
[21:33:38] <deep42thought> :-D
[21:33:49] <abaumann> (feel free to swap Ubuntu for OSX)
[21:34:08] <deep42thought> the only way to *really* remove a bug is to remove the code
[21:34:34] <abaumann> well.. that's rather philosophical.. and drastic. :-)
[21:34:50] <abaumann> mv: cannot stat '/build/libxkbcommon/pkg/libxkbcommon/usr/lib/*x11*': No such file or directory
[21:34:52] <deep42thought> it's a corrolar to the first law of coding which I just mentioned
[21:35:05] <deep42thought> looks like something did not get built ...
[21:35:11] <abaumann> aha. that's another path the maintainer didn't test: everything gets rewritten into some sloppy meson support
[21:36:00] <abaumann> that's the problem with switchs like --enable-X11 (or as a matter of fact with every 'if-then-else')..
[21:36:17] <abaumann> ..you open up a dimension in the testing space
[21:36:24] <deep42thought> it multipies the to-be-tested configurations by 2?
[21:36:41] <abaumann> multiply.. rather compinatorial explosion.
[21:36:49] <abaumann> *combinatorial
[21:36:59] <deep42thought> yeah, exactly
[21:37:10] <abaumann> assuming everything is mutually independent, of course.
[21:37:26] <deep42thought> well, you need to test it anyways
[21:37:37] <deep42thought> even "you cannot have x without y" should be tested
[21:37:42] <abaumann> true
[21:38:25] <abaumann> well. I'm tired...
[21:38:30] <buildmaster> i486/capi4hylafax is broken (says nlopc46-i486bs0): https://archlinux32.org
[21:38:31] <deep42thought> good night :-)
[21:38:33] <abaumann> ..I'll continue (to rant0 tomorrow.
[21:38:34] <abaumann> :-)
[21:38:36] <abaumann> cu
[21:38:39] <deep42thought> cu
[21:38:39] -!- abaumann has quit [Quit: leaving]
[21:43:42] <buildmaster> pentium4/capi4hylafax is broken (says eurobuild6-6): https://archlinux32.org
[21:44:17] <buildmaster> i686/capi4hylafax is broken (says nlopc46): https://archlinux32.org
[21:45:03] -!- isacdaavid has quit [Ping timeout: 272 seconds]
[21:46:25] <buildmaster> pentium4/doxygen is broken (says eurobuild6-6) - already flagged out-of-date upstream on 2020-04-13: https://archlinux32.org
[21:48:04] <buildmaster> i486/qcad is broken (says nlopc46-i486bs0): https://archlinux32.org
[23:17:55] <buildmaster> i686/dmd is broken (says nlopc46): https://archlinux32.org
[23:20:09] -!- deep42thought has quit [Quit: Leaving.]
[23:46:50] <buildmaster> i686/ldc is broken (says eurobuild6-6): https://archlinux32.org
[23:48:16] <buildmaster> pentium4/ldc is broken (says nlopc46): https://archlinux32.org
[23:51:26] <buildmaster> pentium4/dmd is broken (says eurobuild6-6): https://archlinux32.org
[23:53:26] <buildmaster> i686/mustache-d is broken (says nlopc46): https://archlinux32.org
[23:54:39] <buildmaster> pentium4/mustache-d is broken (says eurobuild6-6): https://archlinux32.org
[23:56:39] -!- buildmaster has quit [Remote host closed the connection]
[23:56:55] -!- buildmaster has joined #archlinux32
[23:56:56] <buildmaster> !rq buildmaster
[23:56:56] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)