Switch to DuckDuckGo Search
   February 18, 2020  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | >

Toggle Join/Part | bottom
[00:17:04] *** andy_js <andy_js!~andy@> has quit IRC (Quit: andy_js)
[00:19:39] *** joltman <joltman!znc@gateway/vpn/privateinternetaccess/joltman> has quit IRC (Quit: ZNC - http://znc.in)
[00:20:53] *** joltman <joltman!znc@gateway/vpn/privateinternetaccess/joltman> has joined #illumos
[00:25:40] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[00:28:35] <jbk> rmustacc: i'm suggesting that it might be better to relax it a bit -- having all these bits side pulled (which means we now have at least three different versions floating around -- the illumos-gate one, the omnios one, and the smartos one) doesn't do anyone any good, I would think having that work happen in one place instead of trying to coordinate across three different forks is going to be easier for
[00:28:42] <jbk> everyone.. even if part of it is only used by one of them..
[00:32:08] <rmustacc> jbk: I think if someone has a concrete plan, that's useful.
[00:33:08] <rmustacc> If the answer is there's a consumer in location xyz and we're working on getting that up (and willing to accept the risk of divergence before then), then that's one thing and useful. That's very different from I ahve a bunch of libraries and maybe someone will use them someday in the future, but no one does today and we've never written a single consumer.
[00:33:13] <rmustacc> I think you'd agree those are different?
[00:40:26] <jbk> yes, but in this case there is a downstream consumer.. which today doesn't appear to count...
[00:41:15] <rmustacc> I think the question is is it a downstream consumer that will never come up or not.
[00:41:30] <rmustacc> And I'm trying to say we should be flexible, so work with me?
[00:42:01] <rmustacc> I'm trying to get us to ultimately a yes and not a no.
[00:42:14] <rmustacc> Despite how you seem to be interpetting my rhetoric to the contrary.
[00:42:38] <jbk> the hope is that it gets upstreamed, but it might be a while before that can happen..
[00:43:03] <rmustacc> So, is this just the refhash or something else?
[00:43:06] <jbk> and that there'll be other things along the way..
[00:43:21] <rmustacc> And are you going to be OK that upstream can and will change private interfaces in the interim?
[00:43:26] <jbk> yes just refhash -- and having it live in usr/src/common instead of usr/uts/common/os
[00:43:27] <rmustacc> And use it in ways that break downstream.
[00:44:38] <rmustacc> And as you and andyf pointed out, refhash has a consumer today in at least mpt_sas, so it seems it's not the worst, though kind of weird to move it prematurely.
[00:45:53] <rmustacc> Moving it out of mpt_sas in general is useful. Just some folks will ask why are we building a library we don't ship or if we do, nothing should link against.
[00:47:09] <rmustacc> jbk: So to summarize, let's write down a plan and we can make things work.
[00:47:11] <LeftWing> I think what has been frustrating in the past is the lack of a detailed plan, to be honest
[00:47:13] <LeftWing> Right
[00:47:23] <LeftWing> So like: maybe the plan is going to take a long time
[00:47:28] <LeftWing> Maybe it won't ever complete
[00:47:36] <LeftWing> But we should know what it was when we decided to allow a partial thing
[00:48:00] <LeftWing> Otherwise, future us will look back and wonder why we moved this stuff around, and rightfully might consider moving it back
[00:48:00] <richlowe> That would make it easier to delete the partial thing
[00:48:04] <LeftWing> yes
[00:48:20] <LeftWing> Partial thing probably OK, as long as we can reasonably know when we can decide it didn't work out and abandon it
[00:48:31] <LeftWing> So much of our issues with half-finished stuff in the gate today is because Sun didn't leave us with the plans
[00:48:37] <LeftWing> So we have no idea if it's load bearing or not
[00:48:58] <richlowe> and the challenge historically, and in general, is people do tend to say "I'll do this later" with no intention of doing it later.
[00:49:19] <richlowe> and/or catch a debilitating case of architecture, and start adding infrastructure that doesn't actually hold anything up
[00:49:25] <LeftWing> Which means that for the subset of people who _do_ intend to (and will actually) do it later, they get punished
[00:50:02] <LeftWing> So we could stop punishing the Long Term Plan pattern, as long as there is something written down, basically
[00:50:05] <rmustacc> And in the concrete case that there are downstream things working on it and making changes and want to coordinate and that being illumos helps, that's also very different.
[00:50:39] <LeftWing> Yeah if it's mostly going into the gate so that downstreams can use it (even if just for now) the plan should include some sense of the contract there, I think
[00:50:51] <LeftWing> (Are you expecting we won't move it around while you're not looking, for instance)
[00:51:22] <richlowe> I'm 100% in favour of a phase 1 that's waiting on a phase 2, I'm 100% against a phase 1 where phase 2 is "doing it again, but properly this time"
[00:51:32] <richlowe> the latter being a pattern sun used several times
[00:51:38] <LeftWing> Yeah that's never great
[00:51:47] <LeftWing> Especially in the world where it's kind of murkier what you're allowed to depend on
[00:52:08] *** igork <igork!~igork@> has joined #illumos
[00:52:17] <rmustacc> jbk: Does this help at all?
[00:52:41] <richlowe> it sounds to me like jbk was entirely within the ranges we three seem ok with?
[00:52:56] <LeftWing> Yes, I think so. Writing down how will be the ticket to success, I feel.
[00:53:34] <andyf> For the bhyve upstream work that is planned, there are going to be a fair number of things upstreamed as pre-requisites which won't immediately have consumers
[00:53:39] <LeftWing> Like: where are these other consumers outside the gate, what does the desired end state look like, what's phase 2, etc
[00:53:41] <LeftWing> andyf: Right
[00:53:44] <LeftWing> But we know we want bhyve in there
[00:53:46] <andyf> but I've begun to write it down in an IPD
[00:53:50] <LeftWing> Fantastic!
[00:53:58] <jbk> andyf: have you talked to woodstock at aall?
[00:54:02] <andyf> https://github.com/citrus-it/ipd/tree/master/ipd/0015
[00:54:07] <rmustacc> andyf: I think the main thing is as long as there's a plan, that's what matters.
[00:54:12] <andyf> jbk yes - Woodstock and I are looking at this together
[00:54:31] <LeftWing> Nice
[00:54:47] <jbk> cool.. i know he had said he was working on it..
[00:55:06] <jbk> rmustacc: yes.. if the answer is 'as long as we have a plan' that makes it easier..
[00:55:32] <andyf> Yes, although I think it's in his spare time.. we talked a bit at FOSDEM and he started pulling together commits from omnios
[00:57:25] <rmustacc> jbk: Yes. The thing about the plan is I recommend getting the advocates (or one of them at least) to agree to it so everyone's on the same page in advance.
[00:58:05] <LeftWing> Yeah, that's what I envisage as having a "sponsor" for an IPD; i.e., an Advocate who is Convinced and willing to Go To Bat for it
[00:58:40] <LeftWing> So if you can convince richlowe or rmustacc or myself, I suspect you'll be fine
[00:59:17] *** igork <igork!~igork@> has quit IRC (Quit: Leaving.)
[00:59:39] <rmustacc> And again, we want to say yes and find ways to make things work. Not just be a barrier that shouts no.
[00:59:42] *** igork <igork!~igork@> has joined #illumos
[00:59:55] <LeftWing> Yup
[02:23:46] *** nde <nde!uid414739@gateway/web/irccloud.com/x-ymlutbnpvlnkzkih> has joined #illumos
[03:52:23] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[03:53:11] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[04:02:05] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[04:04:14] <gitomat> [illumos-gate] 12225 normalize complex float names across compilers -- Robert Mustacchi <rm at fingolfin dot org>
[05:06:47] *** _alhazred <_alhazred!~Alex@mobile-access-bceeb6-196.dhcp.inet.fi> has quit IRC ()
[05:56:33] *** leoric <leoric!~leoric@2a02:2698:602b:b48:9513:7735:a223:833d> has joined #illumos
[05:56:52] <leoric> Agnar: Hi, are you here?
[05:57:16] <leoric> Can I ask you to test http://buildzone.oi-build.r61.net/webrev-12316/ ?
[06:19:02] *** tozhu <tozhu!~tozhu@> has quit IRC (Quit: tozhu)
[06:22:27] *** leoric <leoric!~leoric@2a02:2698:602b:b48:9513:7735:a223:833d> has quit IRC (Quit: Konversation terminated!)
[06:25:38] *** tru_tru <tru_tru!~tru@> has quit IRC (Ping timeout: 240 seconds)
[07:11:44] *** tozhu <tozhu!~tozhu@> has joined #illumos
[07:25:16] *** leoric <leoric!~leoric@2a02:2698:602b:b48:1c64:7c53:6f36:ab78> has joined #illumos
[07:29:40] *** cartwright <cartwright!~chatting@gateway/tor-sasl/cantstanya> has quit IRC (Remote host closed the connection)
[07:39:22] *** leoric <leoric!~leoric@2a02:2698:602b:b48:1c64:7c53:6f36:ab78> has quit IRC (Quit: Konversation terminated!)
[07:52:28] *** nde <nde!uid414739@gateway/web/irccloud.com/x-ymlutbnpvlnkzkih> has quit IRC (Quit: Connection closed for inactivity)
[07:56:35] *** tozhu <tozhu!~tozhu@> has quit IRC (Ping timeout: 265 seconds)
[08:00:26] <sjorge> So the aarch64 and risc-v code drops from Hayashi Naoyuki look impressive
[08:00:47] <sjorge> I wonder if this means illumos on raspberry pi 3+ should be doable now, as those are also aarch64
[08:01:39] <v_a_b> sjorge My thoughts exactly. I happen to have an extra Pi 3B here. :-)
[08:24:49] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has quit IRC (Quit: ^C)
[08:33:15] *** neuroserve <neuroserve!~toens@> has joined #illumos
[08:36:43] *** funnyguy <funnyguy!~funnyguy@> has quit IRC (Remote host closed the connection)
[08:37:50] *** funnyguy <funnyguy!funnyguy@2600:3c02::f03c:91ff:fe56:7f51> has joined #illumos
[08:49:04] <sjorge> I wonder if going the uboot->UEFI route like freebsd is doing is a better fit though... then we g et all the loader goodies too
[08:49:15] <sjorge> And I think that allowed them to drop some of the platform specific image for a generic one
[08:49:25] <sjorge> Anyway, thats way ahead of where we are :D
[08:49:38] <sjorge> Looks like they are crosscompiling from linux atm
[08:52:08] <wilbury> crosscompiling what? uefi?
[08:52:35] <tsoome> arm/risc-v
[08:53:01] <wilbury> sjorge: you see, tsoome is one of the most apropriate persons to ask :-)
[08:53:13] *** BOKALDO <BOKALDO!~BOKALDO@> has joined #illumos
[08:53:49] <tsoome> but no, I haven't seen the project bits myself:)
[08:54:36] <wilbury> how came? one could have thought that you are doing all the boot bits
[08:54:38] <wilbury> :-)
[08:55:03] <tsoome> :)
[08:58:30] <tsoome> so far my ability to build and run aarch64 is via qemu, and even building freebsd there (full base) is taking like full week:D
[08:59:21] <tsoome> only building loader bits is much faster obviously
[09:06:57] <sjorge> https://github.com/andreiw/RaspberryPiPkg
[09:15:50] *** leoric <leoric!~alp@pyhalov.cc.rsu.ru> has joined #illumos
[09:18:09] *** BOKALDO_ <BOKALDO_!~BOKALDO@> has joined #illumos
[09:18:32] *** wiedi <wiedi!~wiedi@> has joined #illumos
[09:20:50] *** BOKALDO <BOKALDO!~BOKALDO@> has quit IRC (Ping timeout: 240 seconds)
[09:21:41] <tsoome> LeftWing ping:)
[09:27:40] <tsoome> is there any document for dummies about IPD? Contribution guide does not mention it at all. IPD link is just pointer to GitHub. if some new fancy mechanism is planted, please make sure people actually can get information about it. developing for illumos is already annoyingly complicated. tyvm.
[09:45:30] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[09:53:29] *** Diagon <Diagon!~DiagonalA@unaffiliated/diagonalarg> has joined #illumos
[09:56:29] *** Diagon <Diagon!~DiagonalA@unaffiliated/diagonalarg> has quit IRC (Client Quit)
[09:57:48] *** andy_js <andy_js!~andy@> has joined #illumos
[09:59:41] <sjorge> v_a_b has the person who did the code drop show interest in upstreaming? If so, an IPD might be a good start
[10:12:58] <v_a_b> So far, there was just the one email to illumos-discuss. I am sure that when someone assists him he will agree to upstreaming his work.
[10:16:23] <ptribble> Unfortunately, there wasn't a great deal of interaction last time (with just the aarch64 and alpha ports)
[10:20:28] *** man_u <man_u!~manu@manu2.gandi.net> has joined #illumos
[10:23:31] <jperkin> yeh it's a shame, I think there would have been a lot more traction had it been done natively
[10:40:12] <sjorge> I had a quick sniff and it looks like there it's following the abstraction we had for x86 and sparc and the platforms dirs like we have for sparc
[10:40:16] <v_a_b> This happens all the time. Some people are brilliant, but loners.
[10:40:21] <sjorge> WHich is nice as aarch64 will probably need them.
[10:40:33] <sjorge> Apparently they have been working on it since 2017!
[10:40:57] <sjorge> Can't speak for the code quality though, but it does seem to be pretty up to date (last gate pull was last week)
[11:36:44] <Agnar> leoric: I'm building a new hald over lunch break. let's see if I can test it this afternoon or tonight
[11:40:47] <Agnar> well. if my x230 survives the build ;)
[11:40:51] <Agnar> (thermal)
[11:56:47] *** Agnar <Agnar!~olbohlen@eenfach.de> has quit IRC (Ping timeout: 272 seconds)
[12:07:12] *** Agnar <Agnar!~olbohlen@eenfach.de> has joined #illumos
[12:35:46] *** amrfrsh <amrfrsh!~Thunderbi@> has quit IRC (Quit: amrfrsh)
[13:19:23] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[13:38:58] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:15:51] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[14:19:06] *** amrfrsh <amrfrsh!~Thunderbi@> has joined #illumos
[14:28:45] <ptribble> One for jlevon or tsoome: have those ld guidance changes been tested on SPARC?
[14:28:50] <jlevon> no
[14:29:06] <jlevon> there'll probably be follow up
[14:30:16] <leoric> tsoome: you'll likely like this - tsoome, you'll be trilled to see this - https://gitlab.freedesktop.org/alanc/hal/commit/b841e987de7c91d341fe3a7f8daea36419914503
[14:30:24] <leoric> especially silenced lint
[14:30:50] <ptribble> I'm hoping to get enough time to revisit SPARC builds this week
[14:30:55] <Agnar> leoric: I compiled the patch. will it be enough to replace hald and hal-storage-cleanup-mountpoint?
[14:31:13] <ptribble> so I might discover how much breakage we currently have
[14:31:35] <leoric> likely, however I'd gone for whole /usr/lib/hal
[14:31:45] <Agnar> leoric: ok, yeah, safe bet :)
[14:32:08] <Agnar> leoric: will try that later and give feedback as I is quite re-producable here
[14:33:04] <leoric> we have the same issue, however, not silenced
[14:33:25] <leoric> smatch doesn't find it (even if enabled for hal)
[14:34:40] <gitomat> [illumos-gate] 12233 libsmbios fails to build on SPARC -- Peter Tribble <peter.tribble at gmail dot com>
[14:41:45] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[14:45:47] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Ping timeout: 240 seconds)
[14:48:21] *** tru_tru <tru_tru!~tru@> has joined #illumos
[14:50:28] <tsoome> leoric nice
[14:51:08] <tsoome> ptribble I have "almost" clean build branch pushed on github
[14:51:27] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[14:52:09] <tsoome> only need to double check last bits which i have on either x86 or sparc tree...
[14:53:35] *** hemi770 <hemi770!~hemi666@unaffiliated/hemi770> has quit IRC (Ping timeout: 260 seconds)
[14:54:02] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[14:56:47] *** tsoome_ <tsoome_!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[14:56:50] *** tsoome_ <tsoome_!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Client Quit)
[14:58:16] *** BOKALDO_ <BOKALDO_!~BOKALDO@> has quit IRC (Quit: Leaving)
[15:02:59] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Ping timeout: 272 seconds)
[15:07:57] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has quit IRC (Ping timeout: 268 seconds)
[15:10:31] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[15:27:14] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Read error: Connection reset by peer)
[15:29:46] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[15:35:11] *** hemi770 <hemi770!~hemi666@unaffiliated/hemi770> has joined #illumos
[15:37:11] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 272 seconds)
[15:38:11] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[15:42:41] *** tg2 <tg2!~tg2@> has quit IRC (Ping timeout: 246 seconds)
[15:44:57] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 268 seconds)
[15:46:14] *** BOKALDO <BOKALDO!~BOKALDO@> has joined #illumos
[15:47:26] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[15:58:12] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[16:04:31] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 260 seconds)
[16:12:04] *** wiedi <wiedi!~wiedi@> has quit IRC (Read error: Connection reset by peer)
[16:12:48] *** wiedi <wiedi!~wiedi@> has joined #illumos
[16:15:09] *** ibenn <ibenn!~benn@HSI-KBW-095-208-237-062.hsi5.kabel-badenwuerttemberg.de> has joined #illumos
[16:22:51] <andyf> leoric: ouch, and they even suppressed lint!
[16:24:12] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[16:38:36] *** neuroserve <neuroserve!~toens@> has quit IRC (Ping timeout: 268 seconds)
[16:42:55] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 268 seconds)
[16:52:00] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[16:53:15] <tsoome> andyf try to remove some gags from gate:D
[16:55:06] <rmustacc> tsoome: Yes, I agree we need to improve the documentation there. I can follow up with LeftWing about taking a swing at writing something soon.
[16:55:24] <rmustacc> We need to do a better job of documenting the things we're trying.
[16:57:35] <rmustacc> On the port stuff, if a group of folks are interested in working together on trying to get a bunch of that integrated and working against illumos, I think that'd be worthwhile.
[16:57:51] <rmustacc> Certianly seems like a better starting point than efforts to start from scratch.
[17:00:46] <Woodstock> rmustacc: are you ok with the testing that i did for 9794?
[17:01:54] <rmustacc> Woodstock: Apologies, I missed your reply. I'll circle back on that shortly (about to step afk for transit).
[17:03:37] <tsoome> rmustacc, yea, I mean, if I know nothing about the process and go to check to wiki, it really is not helping me:)
[17:05:32] <rmustacc> tsoome: I understand. I'll put something together there. Same with gerrit.
[17:05:43] <rmustacc> Woodstock: You should be all set.
[17:06:15] <Woodstock> rmustacc: thanks
[17:06:17] <rmustacc> tsoome: I've got some guests in town and some ports to look at, so give me a few days.
[17:06:32] <tsoome> no hurry:D
[17:07:54] <rmustacc> Though one thing I would say is if there's anyone else that's interested in working with the ports that were put up, let me know. I'd like to help organize a group of folks to work on that. Ideally with Hayashi Naoyuki if they're interested. But I also assume they may not be.
[17:08:08] *** nde <nde!uid414739@gateway/web/irccloud.com/x-ctfzyukdpufuciro> has joined #illumos
[17:09:32] <jperkin> I'm still interested in working on the ARM stuff, I think it'll help a lot to be more organised this time so we're on the same setup and can work somewhat independently on things
[17:10:13] *** ZOP <ZOP!~ZOP@phobos.wgops.com> has quit IRC (Ping timeout: 260 seconds)
[17:13:30] <sjorge> Would be nice to pick up on the uboot -> UEFI stuff, that means we could use loader and can skip lots of hte ugly uboot stuff
[17:16:50] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 268 seconds)
[17:18:15] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[17:25:30] *** psarria <psarria!~psarria@37.red-79-146-98.dynamicip.rima-tde.net> has joined #illumos
[17:29:12] *** BOKALDO <BOKALDO!~BOKALDO@> has quit IRC (Read error: Connection reset by peer)
[17:29:17] *** BOKALDO_ <BOKALDO_!~BOKALDO@> has joined #illumos
[17:32:56] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[17:38:02] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 240 seconds)
[17:38:11] <tsoome> sjorge no kidding:D
[17:38:50] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has quit IRC (Remote host closed the connection)
[17:39:17] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has joined #illumos
[17:50:38] <gitomat> [illumos-gate] 12293 libfru: SOLARIS_UNIX is not needed -- Toomas Soome <tsoome at me dot com>
[18:02:10] *** tg2 <tg2!~tg2@> has joined #illumos
[18:08:34] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Quit: man_u)
[18:15:09] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has joined #illumos
[18:21:34] *** ZOP <ZOP!~ZOP@phobos.wgops.com> has joined #illumos
[18:35:39] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Remote host closed the connection)
[18:36:16] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[18:36:25] *** ZOP <ZOP!~ZOP@phobos.wgops.com> has quit IRC (Ping timeout: 272 seconds)
[18:37:44] *** ZOP <ZOP!~ZOP@phobos.wgops.com> has joined #illumos
[18:40:13] *** wiedi <wiedi!~wiedi@> has quit IRC (Ping timeout: 272 seconds)
[18:48:40] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[18:51:52] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Quit: KeiraT)
[18:52:22] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[19:04:54] <gitomat> [illumos-gate] 12311 adjust NATIVE_LIBS in SMF makefiles -- John Levon <john.levon at joyent dot com>
[19:04:55] <gitomat> [illumos-gate] 12312 fix unused lib dependencies -- John Levon <john.levon at joyent dot com>
[19:04:56] <gitomat> [illumos-gate] 12313 nightly should check for ld guidance -- John Levon <john.levon at joyent dot com>
[19:04:57] <gitomat> [illumos-gate] 12314 ld fatal warnings miss some guidance messages -- John Levon <john.levon at joyent dot com>
[19:10:07] *** _alhazred <_alhazred!~Alex@> has joined #illumos
[19:22:51] *** vila <vila!~vila@laubervilliers-659-1-133-94.w80-15.abo.wanadoo.fr> has quit IRC (Remote host closed the connection)
[19:23:26] <toasterson> <freenode_nei "toast: how do I start using Auro"> @freenode_neirac:matrix.wegmueller.it: in the Vagrantfile are the instructions to download task which is a simpler version of make. Then you can run task build to get the binaries.
[19:25:32] *** _alhazred <_alhazred!~Alex@> has quit IRC ()
[19:26:44] *** vila <vila!~vila@laubervilliers-659-1-133-94.w80-15.abo.wanadoo.fr> has joined #illumos
[19:33:50] <wonko> Is there an ansible module to handle stuff like svccfg or do I just need to copy files and run raw commands?
[19:36:34] <tsoome> seems likely, but my google is just as good as yours:) puppet definitely has some support there
[19:41:47] <Agnar> wonko: not svccfg, just svcadm => service module
[19:42:16] <Agnar> wonko: I typically use a jinja2 template and pipe the output into a svccfg import
[19:42:18] <wonko> ok, so copy/command is going to be my best bet there
[19:42:41] <wonko> oh, I didn't know you could pipe, that's nice
[19:43:06] <wonko> have an example of that?
[19:50:57] *** robertdfrench <robertdfrench!~robertdfr@h69-11-139-135.cncrtn.dedicated.static.tds.net> has quit IRC (Ping timeout: 265 seconds)
[19:51:25] <toasterson> wonko (IRC): openindiana modules have support for svccfg
[19:51:46] <toasterson> if you look into oi-userland you can frin them there
[19:55:25] <wonko> toasterson: ok, I'll poke around and see what I find, thanks!
[20:00:11] <gitomat> [illumos-gate] 12185 Remove B100s support -- Peter Tribble <peter.tribble at gmail dot com>
[20:03:36] *** BOKALDO <BOKALDO!~BOKALDO@> has joined #illumos
[20:03:54] *** BOKALDO_ <BOKALDO_!~BOKALDO@> has quit IRC (Read error: Connection reset by peer)
[20:06:45] <toasterson> So any takers to upstream the aarch64 and ricv code drop?
[20:07:40] <ptribble> We're having enough trouble supporting a 2nd architecture in the gate at the moment
[20:09:14] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Ping timeout: 240 seconds)
[20:09:44] <tsoome> we are in trouble supporting even first one:D
[20:11:05] <tsoome> but still it would be really interesting to get at least arm
[20:11:18] <toasterson> Yeah we have a ton of ENORESOURCES
[20:11:25] <tsoome> aarch64 that is.
[20:13:59] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[21:20:40] *** BOKALDO <BOKALDO!~BOKALDO@> has quit IRC (Quit: Leaving)
[21:30:25] *** basvdlei <basvdlei!~basvdlei@2001:980:a4c3:1:b88b:1590:d6bf:fbee> has joined #illumos
[21:33:59] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 268 seconds)
[21:36:27] *** patdk-l2 <patdk-l2!~patrickdk@> has quit IRC (Ping timeout: 240 seconds)
[21:41:30] <tsoome> danmcd, there is is: https://code.illumos.org/c/illumos-gate/+/381
[21:46:26] *** tru_tru <tru_tru!~tru@> has quit IRC (Ping timeout: 240 seconds)
[21:53:47] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #illumos
[21:58:31] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has joined #illumos
[22:10:22] *** dopplergange <dopplergange!~dop@> has quit IRC (Ping timeout: 268 seconds)
[22:13:54] *** neirac <neirac!~cneir@pc-184-104-160-190.cm.vtr.net> has joined #illumos
[22:14:57] *** dopplergange <dopplergange!~dop@> has joined #illumos
[22:16:55] <wonko> Is it just me or is consul running on smartos containers WAY faster than docker containers in Triton?
[22:18:08] <Agnar> leoric: ok, your patch makes it better, but after 5-6 mount/umount using mate desktop, it hangs somewhere and I had to svcadm restart hal. but that is not a typical use case
[22:18:09] <sjorge> native consul will probably be faster, I'd expect
[22:22:04] <wonko> but should there really be that big a difference between native zone and branded zone? That's all sdc-docker does, right? create an LX branded zone from the docker container?
[22:24:37] <neirac> wonko maybe you could check what consul is doing on lxbrand and compare with the native one.
[22:25:07] <ptribble> Ugh. Build failures...
[22:25:20] <wonko> neirac: definitely something to investigate I feel
[22:25:42] <neirac> toasterson I'm interested in open aurora, how do I start working with it?
[22:25:57] <neirac> wonko, you have your triton setup now :)
[22:26:09] <wonko> yep!
[22:26:11] <wonko> it's really nice
[22:37:11] <rmustacc> Well, if we want a second arch to work, it needs to be buildable and runnable from x86.
[22:37:24] <rmustacc> One of the major SPARC problems is no one can boot and test sparc in qemu or similar on x86.
[22:37:38] <rmustacc> For ARM and RISC-V, it can pretty easily be designed to solve that problem.
[22:37:59] <rmustacc> SPARC kind of has the inertia problem of not starting there, which makes it harder.
[22:40:23] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has joined #illumos
[22:43:59] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 265 seconds)
[22:44:47] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has quit IRC (Ping timeout: 260 seconds)
[22:45:42] *** jimklimov1 <jimklimov1!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has quit IRC (Quit: Leaving.)
[22:45:57] <toasterson> neirac (IRC): Have a look at the Vagrantfile for bootstraping and the Taskfile to compile the cli
[22:46:33] <toasterson> It's missing most features though. Only docker pulling and running currently works. not Volumes.
[22:47:00] <neirac> toasterson thanks!
[22:47:06] <toasterson> Latest master may be broken though. I was working on persistent volume support.
[22:48:09] <toasterson> Network naming ist also not final. I am not happy with how vnics are named.
[22:48:21] <toasterson> But it works :)
[22:52:08] <neirac> toasterson I was thinking on using nomad for orchestration and some layer above to create the jobs specs, what's the idea behind opencloud, something like Openshift?
[22:53:55] <andyf> hmm ld: guidance: removal of unused dependency recommended: libmapmalloc.so.1
[22:54:11] <ptribble> ld: guidance: removal of unused dependency recommended: libc.so.1
[22:54:17] <andyf> yeah...
[22:57:00] <ptribble> are there magic runes to turn the ld guidance stuff off?
[22:57:14] <andyf> I think I just need to reverse the order of `-lc -lmapmalloc`
[22:57:15] *** Aram_ is now known as Aram
[22:57:28] <andyf> (this is in lx, so my other option is to wait and see what Jerry does when he next merges :D )
[22:58:47] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 260 seconds)
[23:00:37] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #illumos
[23:00:51] <richlowe> LeftWing: something I missed initially, that awful getdate thing is standard!
[23:02:19] <toasterson> No. More like a layer below Kubernetes or any sheduler (Nomad) that provides secure API's to them to run Container Workloads and create/Shedule nodes on Public/Private Cloud providers as needed to scale the workloads. Bare metal is also a target. Basicly Netboot/Image boot illumos on Something that does Compute for N Containers and manage things like Persistent volumes networks etc for them. The High level concept is that there is
[23:02:20] <toasterson> always a Tennant who has a few resources and a control pane. The controll pane can be on any cloud servers where provisioning is wupported. Linked by either Performant VPN or KCP. Digital Oceans Control Panel is an example how I imagine things to be.
[23:31:59] <neirac> toasterson oh, thanks for the explanation
[23:43:59] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)

   February 18, 2020  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | >