17:59:15 <jcajka> #startmeeting Go SIG meeting
17:59:15 <zodbot> Meeting started Mon Aug  2 17:59:15 2021 UTC.
17:59:15 <zodbot> This meeting is logged and archived in a public location.
17:59:15 <zodbot> The chair is jcajka. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:59:15 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:59:15 <zodbot> The meeting name has been set to 'go_sig_meeting'
17:59:24 <jcajka> hello and welcome all :)
17:59:32 <jcajka> #topic Roll Call
17:59:43 <jcajka> #chair alexsaezm
17:59:43 <zodbot> Current chairs: alexsaezm jcajka
17:59:50 <jcajka> #chair eclipseo
17:59:50 <zodbot> Current chairs: alexsaezm eclipseo jcajka
18:00:59 <alexsaezm> hi there
18:02:52 <jcajka> I hope that summer or winter is going fine for you all. We have one tagged issue in our issue tracker.
18:03:41 <jcajka> That being the deprecated GOPATH
18:04:02 <jcajka> #topic GOPATH deprecated on Go 1.17 https://pagure.io/GoSIG/go-sig/issue/35
18:04:50 <alexsaezm> right, I need to retake that now that we have time. Last time I checked, upstream didn't set a release for the removal of the GOPATH
18:05:43 <jcajka> that will allow more time for solution of it in Fedora
18:05:49 <eclipseo> I believe it was moved to 1.18
18:06:20 <eclipseo> however some big project like k8s don't seem to care about this atall
18:07:40 <alexsaezm> The only tracking issue that I'm aware of targets 1.18 but it only removes a part of the system: https://github.com/golang/go/issues/43684
18:07:57 <jcajka> I think they will react in release that will transition to go1.18, I will be surprised if they move off gopath sooner
18:09:07 <eclipseo> alexsaezm: did you have time to retake a look at nim's previous work
18:09:26 <alexsaezm> no, sorry, I was on a trip
18:10:12 <jcajka> still it seems that it might have nearly no impact on Fedora usage, but still I think sooner for us to move of Gopath the better, at least we can more gracefully transition to the new ways
18:12:24 <eclipseo> yes, the newpossibility regarding packaging are interesting, although it will be quite the workflow change if we follow nim initial macros. No more tests deps separated from normal deps, test not being bypassable from the commnd line either
18:13:41 <jcajka> yeah, I think we should strive to have these kept, but that means to have more tooling on our side or at least needed to be upstreamed in some way
18:14:52 <eclipseo> upstream is not really listening to our requests, they don't seem to care about downstream problems
18:15:22 <alexsaezm> eclipseo, why you say that? did I miss something?
18:15:28 <jcajka> if I haven't missed anything it looked more they don't listen to nim
18:16:14 <jcajka> but we will have to find some way or more precisely someone who will pick this work up
18:16:27 <eclipseo> forexample nim tried to ask them multiples times to have a way to differentiate between test import paths and normal import path, and it has fallenon deaf ears
18:18:22 <jcajka> I hope that is was more due to more confrontational style of nim than anything else
18:18:37 <eclipseo> idk
18:19:24 <alexsaezm> if that is necessary, we can try to move that again
18:19:39 <eclipseo> https://github.com/golang/go/issues/26955
18:21:26 <eclipseo> i hadn't see this message: https://github.com/golang/go/issues/29410#issuecomment-515203754
18:21:53 <alexsaezm> thanks for the link
18:23:31 <jcajka> eclipseo: interesting new notes, thanks :)
18:25:46 <eclipseo> alexsaezm: will you have more time to dedicate to this issue?
18:26:01 <alexsaezm> yes
18:26:06 <alexsaezm> for sure
18:26:28 <alexsaezm> last months weren't that good tbh but I have time for this
18:27:40 <eclipseo> great thanks
18:27:47 <jcajka> great :)
18:28:59 <jcajka> seems that we can move to the last topic
18:30:12 <eclipseo> ok
18:30:41 <jcajka> to the open floor
18:30:55 <jcajka> #topic Open Floor
18:31:53 <eclipseo> ok, so status report from me: i am taking care of my FTBFS, it's progressing slowly
18:32:25 <jcajka> eclipseo: nice, any significant issue, out liners?
18:32:38 <jcajka> I can take a look
18:33:35 <jcajka> that bring me to one that you might see, but it shouldn't happen in koji, due to older kernels there
18:33:57 <eclipseo> no major issue so far, just some juggling with libraries incompabilities as usual, since our policy is to package the latest version
18:34:14 <jcajka> good to hear that
18:34:28 <jcajka> there seems to be significant compiler bug on ppc64le, that might warrant partial rebuild for all releases of Fedora if there is no other way to fix it, workaround
18:34:43 <jcajka> https://bugzilla.redhat.com/show_bug.cgi?id=1987972
18:35:16 <jcajka> fixes for the compiler are in Fedora already, but if any package crashes it unfortunately needs to be rebuild
18:37:05 <eclipseo> ok i hadn't noticed that yet
18:39:20 <jcajka> ositive side is it is only ppc64le and it shouldn't show in koji as it is currently running 5.12, not blocking whole distribution yet
18:39:41 <jcajka> will work with alexsaezm and other hopefully solving it soon
18:40:29 <alexsaezm> sure :)
18:40:41 <jcajka> and that is all from myside
18:41:29 <eclipseo> me too
18:41:37 <alexsaezm> nothing from my side
18:42:27 <jcajka> I guess this will be shorter meeting :)
18:42:38 <jcajka> #endmeeting