<div class="gmail_quote"><div>MINUTES: OpenEmbedded Technical Steering Committee, 24 March 2011</div><div><br></div><div>Attendees: Tom, Khem, Koen, Stefan, Richard</div><div>Apologies: Mark</div><div>Secretary: Jefro</div>
<div><br></div><div> 01) Agree on meeting chair (Khem had previously volunteered)</div><div>___________________________________</div>
<div> 02) Status report on oe-core</div><div><br></div><div>process going well, developers migrating</div><div>Khem suggests need to accelerate meta-oe to bring more users over</div><div>Koen still pushing items from meta-oe to oe-core</div>
<div>Koen: now 3 oe-core distros: poky, angstrom, and shr</div><div>Khem suggests creating separate repos for different distros</div><div>biggest barrier to entry is lack of documentation</div><div>RP & Khem discussed documentation thoughts, more needed</div>
<div><br></div><div>action items:</div><div> general: encourage more distros to participate in oe-core</div><div> general: invite more people to meta-oe on the oe-devel mailing list</div><div> interested parties: discuss documentation</div>
<div><br></div><div>previous action items:</div><div>Koen to send email gathering input for checklist (sent before meeting ended)</div><div>//done</div><div>Jefro volunteers to wikify checklist if someone sends</div><div>
// not done yet - no response on mailing list</div><div>discussion for branding & recipe split to mailing list, Richard responded to Koen</div><div>// proceeding as shown above</div><div>___________________________________</div>
<div> 03) Status report on pull model, contrib repo and guidelines</div><div><br></div><div>Mark's draft commit/patch guidelines well received, not yet finalized</div><div>still need to wikify</div><div><br></div><div>
action item:</div><div> khem will try to write up how to use contrib repos</div><div> </div><div>previous action items:</div><div>Mark hopes to post commit/patch policy guidelines tomorrow, has feedback to incorporate</div>
<div>// not yet done, but soon - action item continued</div><div>Koen would like to copy oe-core guidelines as much as possible for meta-oe layer</div><div>// pending on above</div><div>___________________________________</div>
<div> 04) Status report on board support layer guidelines</div><div><br></div><div>previous action items:</div><div>Richard has action to write proposal for YP steering group, won't have answer until their meeting</div>
<div>// YP steering group meeting next week - action item continued</div><div>___________________________________</div><div> 05) Status report on version retention policy and interaction with</div><div> oe-core / meta-oe / $distro layers</div>
<div><br></div><div>previous action items:</div><div>Tom will wiki it</div><div>// not yet - action item continued</div><div>___________________________________</div><div> 06) Status on layer splitting of metadata</div><div>
<br></div><div>meta-efl created a discussion</div><div>better layer tooling needed</div><div><br></div><div>action items:</div><div> general: keep discussions going</div><div> RP: will take notes of experiences and needs</div>
<div> </div><div>previous action items:</div><div>Koen volunteers to be pullmaster</div><div>Koen will move data in repo down a level into meta-oe now</div><div>//done</div><div>___________________________________</div>
<div>
07) DISTRO_FEATURE/MACHINE_FEATURE/libc features and other</div><div> "flags" discussion</div><div> [ Proposed: Tom Rini ]</div><div><br></div><div>Some discussion on ml</div><div><br></div><div>action item:</div>
<div> all: discuss on mailing list</div><div><br></div><div>previous action items:</div><div>Tom to move to ml</div><div>//done</div><div>___________________________________</div><div> 08) Continue discussion on posting of policies and guidelines</div>
<div><br></div><div>previous action items:</div><div>AI: Khem to highlight the oe-core ml on oe-dev, tell people about discussion</div><div>//done, no response yet</div><div><br></div><div>___________________________________</div>
<div> 09) Continue discussions on infrastructure items</div><div><br></div><div>Bugzilla is dead, meaning it has been taken offline</div><div>Khem added message directing people to use the mailing list for bugs</div><div>
Tom K will work on making database RO</div><div><br></div><div>action item:</div><div> Jefro: write a blurb describing the change for a TSC member to post</div><div><br></div><div>_______________________________________________________________</div>
<div>TRANSCRIPT:</div><div><br></div><div><div><br></div><div>(1:07:12 PM) khem: so item 1. Where do we stand with oe-core</div><div>(1:07:36 PM) koen: agenda: <a href="http://pastebin.com/51VBKfMk" target="_blank">http://pastebin.com/51VBKfMk</a></div>
<div>(1:07:41 PM) RP__: Good question. How are people finding things at the moment?</div><div>(1:07:51 PM) Jefro1: koen - thanks</div><div>(1:08:01 PM) khem: <a href="http://pastebin.com/zWCWCPHg" target="_blank">http://pastebin.com/zWCWCPHg</a> is agenda</div>
<div>(1:08:07 PM) khem: oh thank koen</div><div>(1:08:09 PM) stefan_schmidt: RP__: I think the progress is good</div><div>(1:08:21 PM) stefan_schmidt: faster reaction on patches and more people contributing</div><div>(1:08:24 PM) koen: gettting more OE hackers on board (e.g. Martin)</div>
<div>(1:08:37 PM) RP__: From where I am things are looking reasonable</div><div>(1:09:04 PM) khem: yes I think we have to accelerate meta-oe to bring more users over</div><div>(1:09:08 PM) RP__: Reaction time should be resolved now I have some workflow and the yocto connection is established and rolling</div>
<div>(1:09:16 PM) RP__: More users does seem like the next move</div><div>(1:09:23 PM) koen: I'm still doing a lousy job at pushing things from meta-oe to oe-core, but the diff is getting less</div><div>(1:09:27 PM) stefan_schmidt: khem: agreed, thats what is missing</div>
<div>(1:09:50 PM) RP__: I am trying to give patches some time on the mailing list for comments before merging them where I think its appropriate</div><div>(1:10:06 PM) khem: yes thats right thing to do</div><div>(1:10:06 PM) ***stefan_schmidt thinks many people are in waiting position before doing real work on oe-core</div>
<div>(1:10:12 PM) RP__: koen: I can't talk given the bitbake situation, but likewise, its getting better I think</div><div>(1:10:25 PM) RP__: fray sorted a key pseudo related piece recently</div><div>(1:10:36 PM) koen: there's meta-shr now</div>
<div>(1:10:47 PM) koen: so there are 3 distros for oe-core now: poky, angstrom and shr</div><div>(1:10:55 PM) khem: koen: meta-oe is a good testing pad before pushing oe changes into oe-core</div><div>(1:11:00 PM) RP__: I helped fix a blocking issue on meta-shr so it should be working better now</div>
<div>(1:11:03 PM) koen: <a href="http://git.shr-project.org/git/?p=meta-shr.git;a=summary" target="_blank">http://git.shr-project.org/git/?p=meta-shr.git;a=summary</a></div><div>(1:11:18 PM) khem: I am working on getting uclibc-git/0.9.32/nptl into meta-oe</div>
<div>(1:11:37 PM) koen: as said above, we just need more people to work on oe-core and meta-oe</div><div>(1:11:46 PM) stefan_schmidt: ok, so we all see progress</div><div>(1:11:49 PM) khem: I wanted to encourage other distros on oe to use oe-core</div>
<div>(1:11:57 PM) stefan_schmidt: any concrete steps for more users</div><div>(1:11:58 PM) khem: e.g. slugos kaeilos</div><div>(1:12:23 PM) RP__: khem: remember that slugos focuses more on stability</div><div>(1:12:26 PM) koen: khem: kaeios is easy, just cp meta-angstrom meta-kaailos and s/angstrom/kaeilos/g</div>
<div>(1:12:47 PM) khem: RP__: yes oe-core is just fit for slugos</div><div>(1:13:07 PM) khem: koen: Should we create separate repos for different distros ?</div><div>(1:13:12 PM) RP__: khem: so they might prefer to move around a release stabilisation point rather than now when we're in a bit more of a development mindset</div>
<div>(1:13:13 PM) Tartarus: khem: yes</div><div>(1:13:19 PM) khem: well its maintainers call where they want to host it though</div><div>(1:13:19 PM) Tartarus: distros are meta-distroname</div><div>(1:13:32 PM) khem: RP__: ok</div>
<div>(1:13:33 PM) Tartarus: Unless we want to encourage distros within meta-oe, which i don't know if we do</div><div>(1:13:36 PM) Tartarus: did we talk about that?</div><div>(1:13:41 PM) koen: yes, we don't want that</div>
<div>(1:13:47 PM) khem: I think distros are clearly a separate layer</div><div>(1:13:57 PM) RP__: I don't mind if a distro wants to create a layer there with some policy</div><div>(1:14:09 PM) RP__: equally I don't mind the separation</div>
<div>(1:14:16 PM) khem: hmm like kaelos is mostly angstrom</div><div>(1:14:36 PM) RP__: What I care about is clear ownership and namespace</div><div>(1:14:48 PM) khem: I think micro has some uniqueness but not sure about minimal</div>
<div>(1:15:06 PM) koen: khem: he angstrom devs have a betting pool how soon Marco rips off changes from angstrom after each commit we do</div><div>(1:15:09 PM) Tartarus: Lets set micro/minimal aside for now :)</div><div>
(1:15:13 PM) stefan_schmidt: ok, so we try to encourage other distros</div>
<div>(1:15:27 PM) khem: yes</div><div>(1:15:47 PM) stefan_schmidt: anything else for 1?</div><div>(1:15:48 PM) Jefro1: is there an action item for any particular person on this?</div><div>(1:15:54 PM) koen: personally, I would say that if a distro needs encouraging at this point it doesn't bode well forthe future</div>
<div>(1:16:31 PM) Tartarus: Well, i dunno koen</div><div>(1:16:33 PM) khem: OE has a quilt patch where it create symlinks to patches instead of copying them</div><div>(1:16:34 PM) Tartarus: Here's a maybe AI</div><div>
(1:16:50 PM) khem: I was thinking of bringing that in if its useful</div><div>(1:16:55 PM) Tartarus: How much longer until we want to do an open call or whatever of "Hey, come get over in oe-ore/meta-$whatever and kick the tires folks!"</div>
<div>(1:17:05 PM) Tartarus: We still don't have docs for getting started, do we?</div><div>(1:17:07 PM) RP__: khem: I thought we did that in oecore already? :/</div><div>(1:17:16 PM) khem: RP no</div><div>(1:17:25 PM) khem: and oe-core patches dont rewind properly</div>
<div>(1:17:29 PM) khem: if a patch failes</div><div>(1:17:35 PM) khem: in some cases</div><div>(1:17:36 PM) RP__: Tartarus: Perhaps the gating factor at the moment is the docs?</div><div>(1:17:39 PM) Tartarus: I think we've probably got folks sitting on the side waiting for a "we think the core is usable enough now to start poking for everyone"</div>
<div>(1:18:05 PM) Tartarus: RP__: docs and an open invitation that folks should expect something to work out of the box, that's in the docs, yes</div><div>(1:18:13 PM) khem: I think biggest problem for users I am envisioning is where to start</div>
<div>(1:18:20 PM) Tartarus: khem, yeah</div><div>(1:18:24 PM) Tartarus: It used to just be bitbake nano</div><div>(1:18:32 PM) Tartarus: but that howto needs an update or two, for oe-core/etc</div><div>(1:18:38 PM) Jefro1: I suggest action item: document how to get started with oe-core on the wiki, and announce to the list</div>
<div>(1:18:43 PM) khem: so different distros should take and AR to spread the word</div><div>(1:19:21 PM) RP__: I think that is a good start</div><div>(1:19:34 PM) RP__: Lets try and feed this a little gradually rather than have a thundering herd :)</div>
<div>(1:19:44 PM) khem: ok</div><div>(1:20:04 PM) koen: Let's start with inviting people to meta-oe</div><div>(1:20:11 PM) koen: they can kick that tire :)</div><div>(1:20:31 PM) khem: should we add a new ml for meta-oe</div>
<div>(1:20:46 PM) khem: or use existing oe-devel with properl subject prefix</div><div>(1:21:02 PM) RP__: I suspect oe-devel is the right list for this</div><div>(1:21:09 PM) stefan_schmidt: I agree</div><div>(1:21:11 PM) khem: ok</div>
<div>(1:21:12 PM) Tartarus: Yeah, we talked before</div><div>(1:21:14 PM) Tartarus: no sep list</div><div>(1:21:23 PM) stefan_schmidt: If it gets to much we can still start new list</div><div>(1:21:26 PM) khem: lets use [meta-oe] prefix to make it clear</div>
<div>(1:21:36 PM) khem: although patchwork would not like it</div><div>(1:21:48 PM) koen: the readme says oe-core: Chttp://<a href="http://cgit.openembedded.org/cgit.cgi/meta-openembedded/tree/meta-oe/README" target="_blank">cgit.openembedded.org/cgit.cgi/meta-openembedded/tree/meta-oe/README</a> :)</div>
<div>(1:21:49 PM) RP__: I think if people start posting patches there and pulling them it will get attention in its own right too</div><div>(1:21:54 PM) koen: (that can be changed of course)</div><div>(1:22:10 PM) RP__: Lets fix the README then :)</div>
<div>(1:22:24 PM) khem: ok so koen sent email for getting checklist</div><div>(1:22:34 PM) khem: what do we have on that subject</div><div>(1:22:58 PM) ***khem is still on item 2 sub bullets</div><div>(1:24:17 PM) khem: Jefro1: did you get checklist compiled</div>
<div>(1:24:34 PM) RP__: Mark knows he has things to do and in his email said he'd get to them soon and appologised</div><div>(1:24:42 PM) khem: ok</div><div>(1:24:49 PM) Jefro1: khem - not yet</div><div>(1:25:00 PM) khem: ok we keep it open. Next is branding</div>
<div>(1:25:24 PM) khem: is there a preferece ?</div><div>(1:25:28 PM) koen: I still haven't made the time to rename some poky bits</div><div>(1:25:45 PM) khem: ok so we will look upto you koen</div><div>(1:26:07 PM) khem: anything on recipe split ?</div>
<div>(1:26:13 PM) koen: I don't use any of the poky scripts TBH</div><div>(1:26:21 PM) khem: right</div><div>(1:26:31 PM) RP__: So does poky take them separately?</div><div>(1:26:50 PM) stefan_schmidt: so branding is not such a high prio and we are just waiting for patches wrt this?</div>
<div>(1:26:58 PM) koen: I still think sato goes into meta-oe, since we have it in OE as well, but I'm open to suggestions</div><div>(1:26:59 PM) RP__: One topic I've been thinking about is documentation too. I really want to split off the yocto docs into a separate repo</div>
<div>(1:27:41 PM) khem: hmm so documentation will be available on yocto only</div><div>(1:27:54 PM) RP__: khem: well, depends what we do and how</div><div>(1:27:54 PM) khem: since yocto will have different policies ?</div>
<div>(1:28:07 PM) RP__: khem: The trouble is there are large chunks which are not oecore</div><div>(1:28:21 PM) khem: RP__: I see yes</div><div>(1:28:38 PM) khem: so may be divide that and remove the non oecore</div><div>
(1:29:02 PM) khem: and create a separate yocto-docs repo upto you</div><div>(1:29:15 PM) khem: should we assign an AR to RP__</div><div>(1:29:43 PM) RP__: The other trouble is the current git merge issues are a nightmare for our tech writer</div>
<div>(1:30:00 PM) khem: I see</div><div>(1:30:23 PM) khem: but one feature of oecore should be "its well documented"</div><div>(1:31:03 PM) khem: I would say dividing it into two would be one way</div><div>(1:31:27 PM) RP__: khem: There are two competing goals - document OECore and document Poky (the combined set of things)</div>
<div>(1:31:45 PM) khem: I understand</div><div>(1:31:51 PM) RP__: If you restrict one manual to only talk about OECore, it makes writing flowing documentation hard</div><div>(1:32:32 PM) RP__: I don't have a good answer here yet, I suspect its not as simple as splitting into two</div>
<div>(1:32:32 PM) khem: you can have two volumes one describing OE core another describing yocto/poky may be</div><div>(1:32:43 PM) khem: I guess yes</div><div>(1:33:13 PM) RP__: khem: Its thing like when you want to talk about distro config example files</div>
<div>(1:33:19 PM) khem: but if oe-core is described as oe-core is in a separate repo should be ok as well I would think</div><div>(1:33:57 PM) RP__: khem: I did think this would be much easier than it is before I tried to split it up :/</div>
<div>(1:34:02 PM) Jefro1: documenting separate projects that intertwine is a known issue for a technical writer - but yocto only has one, resources highly limited</div><div>(1:34:35 PM) Jefro1: RP__ I'd be happy to discuss docs offline</div>
<div>(1:34:59 PM) khem: so I guess we need to separate out</div><div>(1:35:03 PM) RP__: Lets not rathole on this now but yes, more discussion needed</div><div>(1:35:04 PM) khem: ok</div><div>(1:35:06 PM) ***stefan_schmidt points out that we only have 26 min left for 5 items</div>
<div>(1:35:24 PM) khem: we are onto 3</div><div>(1:35:41 PM) khem: did Mark post the commit policy</div><div>(1:35:53 PM) RP__: I think we've seen a version?</div><div>(1:36:01 PM) RP__: Not sure it was the final one?</div>
<div>(1:36:04 PM) stefan_schmidt: it was well received</div><div>(1:36:05 PM) Jefro1: there is a draft floating around, I don't think he has put it on the wiki</div><div>(1:36:15 PM) khem: ok lets keep that open</div>
<div>(1:37:07 PM) stefan_schmidt: For the pull model: I think it works well for oe-core and 2011.03 so far</div><div>(1:37:13 PM) khem: do we have more on guidelines for pull model and repo</div><div>(1:37:40 PM) Tartarus: I think we still need to wikify whatever we're doing</div>
<div>(1:37:43 PM) RP__: khem: I think this is what fray has been working on?</div><div>(1:37:45 PM) stefan_schmidt: not sure about contrib repos as koen and /me are using external git servers for our branches</div><div>(1:37:59 PM) RP__: We do have a contrib repo there...</div>
<div>(1:38:02 PM) Tartarus: stefan_schmidt: They're up and working, and optional</div><div>(1:38:09 PM) RP__: For 4), the YP steering meeting is scheduled for early next week and I will have a proposal for them</div>
<div>
(1:38:13 PM) Tartarus: khem and I (and martin and chris larson and others) have been using it</div><div>(1:38:29 PM) khem: RP__: was that for item 04 ?</div><div>(1:38:41 PM) stefan_schmidt: cool, I'm fine with the contrib repos. Was just wondering who is using them</div>
<div>(1:38:42 PM) RP__: khem: yes</div><div>(1:38:55 PM) RP__: stefan_schmidt: I post patches there too</div><div>(1:38:56 PM) khem: Tartarus: we need to document how-to use contrib repo somewhere</div><div>(1:38:57 PM) Tartarus: Might be nice to make it easier for create-pull-request to reference an external repo, but that's minor I think</div>
<div>(1:39:01 PM) Tartarus: khem, yes</div><div>(1:39:10 PM) RP__: Tartarus: send patches ;-)</div><div>(1:39:17 PM) Tartarus: RP__: on my list, heh</div><div>(1:39:23 PM) RP__: Tartarus: I've talked to others and we want to make it configurable</div>
<div>(1:39:37 PM) khem: yes may be a env var or somesuch</div><div>(1:39:44 PM) Tartarus: Anyhow</div><div>(1:39:59 PM) khem: I will try to do a writeup on using contrib repos</div><div>(1:40:00 PM) RP__: Tartarus: user defaults, then repo defaults, then script fall back</div>
<div>(1:40:15 PM) Tartarus: khem: k</div><div>(1:40:27 PM) koen: are we ready for 05)?</div><div>(1:40:28 PM) khem: ok moving on 04 is covered unless someone has more inputs atm</div><div>(1:40:44 PM) khem: moving to 05 then</div>
<div>(1:40:55 PM) khem: version retention policy</div><div>(1:41:15 PM) koen: I think we all agreed and want it in the wiki</div><div>(1:41:38 PM) khem: ok is someone owning to document it there</div><div>(1:41:52 PM) khem: ah tom</div>
<div>(1:41:53 PM) stefan_schmidt: Tartarus is iirc</div><div>(1:41:58 PM) Tartarus: I own it, yes</div><div>(1:42:02 PM) stefan_schmidt: ah, its even in the agenda :)</div><div>(1:42:07 PM) Tartarus: So AI to me, put it on the wiki</div>
<div>(1:42:14 PM) Tartarus: suggested page name?</div><div>(1:42:20 PM) Tartarus: Just Version_Retention_Policy ?</div><div>(1:42:27 PM) khem: Version Retetion Policy</div><div>(1:42:28 PM) stefan_schmidt: sounds good</div>
<div>(1:42:36 PM) khem: with spellcheck</div><div>(1:42:36 PM) Tartarus: k</div><div>(1:42:52 PM) khem: moving on to 06</div><div>(1:43:07 PM) khem: layer splitting</div><div>(1:43:17 PM) koen: the meta-efl thing kicked off a discussion</div>
<div>(1:43:25 PM) khem: yes</div><div>(1:43:33 PM) RP__: it did :)</div><div>(1:43:37 PM) koen: I think we can continue discussion in that thread</div><div>(1:43:41 PM) khem: ok</div><div>(1:43:50 PM) khem: and report next week of outcome</div>
<div>(1:43:51 PM) RP__: Was there any fundamental disagreement in that thread though?</div><div>(1:44:02 PM) khem: I think no</div><div>(1:44:04 PM) koen: RP__: not AFAIK</div><div>(1:44:05 PM) stefan_schmidt: I don't think so</div>
<div>(1:44:26 PM) stefan_schmidt: but it raises the prio for better layer tooling to keep all the blocks together</div><div>(1:44:34 PM) RP__: right, just checking I wasn't missing something :)</div><div>(1:44:43 PM) RP__: agreed, we need better layer toolcing</div>
<div>(1:44:54 PM) Tartarus: We need to note up what we need in tools</div><div>(1:45:04 PM) Tartarus: Best practices and all of that too</div><div>(1:45:09 PM) ***RP__ notes Yocto 1.1 planning is open and layer tooling is on the list</div>
<div>(1:45:24 PM) RP__: but yes, we need to note requirements like this and come up with some kind of plan</div><div>(1:45:24 PM) khem: may be a bill of recipes from everylayer</div><div>(1:45:28 PM) stefan_schmidt: RP__: Seen it. Is it your task or someone else?</div>
<div>(1:45:57 PM) RP__: stefan_schmidt: I own it from an architecture point of view but I'm very open to ideas, discussion and help</div><div>(1:46:09 PM) stefan_schmidt: RP__: ok</div><div>(1:46:21 PM) khem: RP__: OK so I guess we can keep discussions going and gather experiences with layers</div>
<div>(1:46:24 PM) RP__: I don't think it will be a solved overnight problem but we need a plan</div><div>(1:46:28 PM) khem: I guess its premature as of now</div><div>(1:46:41 PM) RP__: I'm happy to take notes of experiences and needs</div>
<div>(1:46:42 PM) stefan_schmidt: sure, we need to learn what we need while going</div><div>(1:47:03 PM) khem: so moving to 07</div><div>(1:47:22 PM) khem: Tartarus: on features</div><div>(1:47:26 PM) stefan_schmidt: oh, the always loved topic :)</div>
<div>(1:47:30 PM) Tartarus: OK</div><div>(1:47:36 PM) Tartarus: Started discussion on ML, later than I wanted</div><div>(1:47:41 PM) khem: ok</div><div>(1:47:43 PM) Tartarus: waiting for real discussion and I assume folks are busy</div>
<div>(1:47:55 PM) khem: any feedbacks you would like to discuss</div><div>(1:48:20 PM) Tartarus: Looking for more details from RP__ and fray who I think have both thought about this before</div><div>(1:48:40 PM) khem: I think posix is a good diving point</div>
<div>(1:48:43 PM) RP__: Tartarus: Ah, yes. Sorry, I've not had time to start writing down things on that subject</div><div>(1:48:44 PM) khem: from libc pov</div><div>(1:49:00 PM) Tartarus: khem, so you're thinking on the libc config point</div>
<div>(1:49:04 PM) khem: yes</div><div>(1:49:05 PM) Tartarus: Yeah, also needs discussion</div><div>(1:49:16 PM) Tartarus: but to me, that's a sub-point of the big problem of controlling what we do and don't build</div>
<div>(1:49:28 PM) RP__: to me too</div><div>(1:49:32 PM) koen: I've been thinking about how to use a checksum for it all</div><div>(1:49:43 PM) Tartarus: OK, so I'm waiting for everyone to chime in on the ML</div>
<div>(1:49:46 PM) Tartarus: please do :)</div><div>(1:49:55 PM) RP__: The checksums already capture any configuration information</div><div>(1:50:09 PM) koen: for me a build/checksum/tmp would be a good start</div><div>(1:50:10 PM) khem: chris was doing a patch for bb</div>
<div>(1:50:18 PM) khem: to use checksumming for recipes</div><div>(1:50:19 PM) koen: change somethign and it's a rebuild in a different dir</div><div>(1:50:36 PM) RP__: koen: which somethings though? :)</div><div>(1:51:19 PM) khem: I think we should keep it churning on ml for a while</div>
<div>(1:51:19 PM) RP__: Anyhow, I think we need to take this up on the mailing list</div><div>(1:51:47 PM) khem: for my AI I did send emails to all mls informing about oe-core</div><div>(1:52:38 PM) RP__: khem: much response?</div>
<div>(1:52:48 PM) khem: RP__: no response</div><div>(1:52:55 PM) RP__: khem: :/</div><div>(1:53:03 PM) khem: RP__: I send it to oe-users list</div><div>(1:53:04 PM) khem: too</div><div>(1:53:13 PM) RP__: Has anyone looked much at oe-lite btw?</div>
<div>(1:53:19 PM) koen: no</div><div>(1:53:21 PM) Jefro1: it is hard to get people excited about policies & guidelines</div><div>(1:53:21 PM) khem: I guess it did its part of informing</div><div>(1:53:33 PM) khem: I have not looked at oe-lite</div>
<div>(1:53:52 PM) RP__: Esben keeps mentioning it and wanting to merge into OE</div><div>(1:54:11 PM) koen: RP__: the unpaid yocto employee remark cracked me up</div><div>(1:54:15 PM) khem: I guess we should encourange him to port it over to some extent</div>
<div>(1:54:23 PM) khem: so we can see benefits</div><div>(1:54:28 PM) RP__: The trouble is he radically "simplified" things by removing things like package management :/</div><div>(1:54:41 PM) khem: yeah</div><div>
(1:54:45 PM) koen: RP__: recipe based sysroots create a huge potential for rootfs breakage</div><div>(1:54:58 PM) RP__: koen: yes, advantages and disadvantages</div><div>(1:54:59 PM) koen: RP__: e.g. libnl1 and libnl2</div>
<div>(1:55:29 PM) khem: I think if we cache complete autoconf vars that will be helpful in deterministic builds</div><div>(1:55:35 PM) RP__: koen: If we ever do recipe bases sysroots, they need to be optional</div><div>(1:55:58 PM) koen: RP__: if we do them, the distro becomes even more important :)</div>
<div>(1:56:16 PM) RP__: khem: there are several ways to approach that problem. The correct approach is to use various approaches I think :)</div><div>(1:56:33 PM) RP__: (swabber, recipe specific sysroot, autoconf cache etc)</div>
<div>(1:56:39 PM) Tartarus: So, what item is this on the agenda?</div><div>(1:56:52 PM) RP__: Tartarus: We're derailing slightly :/</div><div>(1:56:53 PM) koen: RP__: is running tasks under qemu on the 1.1 list?</div>
<div>(1:56:54 PM) khem: I will read that thread</div><div>(1:56:55 PM) RP__: my fault, sorry</div><div>(1:56:58 PM) Tartarus: :)</div><div>(1:57:01 PM) Tartarus: So, next AI?</div><div>(1:57:10 PM) khem: Infra</div><div>
(1:57:16 PM) RP__: koen: not really, we've been trying to escape that</div>
<div>(1:57:23 PM) koen: 8) Continue discussions on infrastructure items</div><div>(1:57:30 PM) koen: RP__: ok</div><div>(1:57:37 PM) koen: bugzilla is dead</div><div>(1:57:41 PM) khem: on infra I added a message to bugzilla to direct people to use ml</div>
<div>(1:57:56 PM) RP__: koen: no reason it couldn't be on there if someone wanted it for a good reason but its not there atm</div><div>(1:57:57 PM) khem: and Tom K will work on making database ro</div><div>(1:58:02 PM) koen: Jefro1 will write some blurb, but we need to send it</div>
<div>(1:58:12 PM) Jefro1: yep</div><div>(1:58:27 PM) RP__: We're going to get hate mail about the bugzilla, people know that, right?</div><div>(1:58:31 PM) koen: RP__: introspection (we'll dicuss on the ml further)</div>
<div>(1:58:38 PM) khem: should we have oe-core bugs reported in yocto bugz</div><div>(1:58:52 PM) ***RP__ mutters darkly about introspection</div><div>(1:59:02 PM) koen: RP__: the question is, will we get hatemail from people that actually matter to the project?</div>
<div>(1:59:25 PM) Tartarus: I asked for and got 72h after saying we need volunteers</div><div>(1:59:31 PM) khem: I think I was maintaining bugz for past years and have no more appetite</div><div>(1:59:31 PM) Jefro1: RP__ re hate mail - why do you think I opted out of posting the notice? ;)</div>
<div>(1:59:32 PM) Tartarus: and got none</div><div>(1:59:32 PM) Tartarus: So</div><div>(1:59:58 PM) RP__: I do think its fair to do this, I just want to ensure people realise the implications</div><div>(2:00:06 PM) khem: we have one min left</div>
<div>(2:00:11 PM) RP__: Someone needs to make a decision and that is us</div><div>(2:00:18 PM) RP__: and I think its the right decision</div><div>(2:00:27 PM) khem: We are not deleting it anyway</div><div>(2:00:31 PM) RP__: Re: OECore bugs in the yocto bugzilla, I'm ok with that</div>
<div>(2:00:42 PM) RP__: as long as its clear they are OECore bugs, not meta-oe</div><div>(2:00:57 PM) khem: that will be hard</div><div>(2:01:07 PM) khem: I think you cant stop folks</div><div>(2:01:25 PM) khem: and some wont be able to make judgement</div>
<div>(2:01:33 PM) khem: so you need to triage them hard</div><div>(2:01:34 PM) RP__: khem: Well, bugs not in OECore/poky will just get closed as invalid</div><div>(2:01:53 PM) RP__: I'm happy to do that, I just want to make it clear what the expectation is</div>
<div>(2:02:18 PM) Jefro1: RP__ out of curiosity, how many bugs in the db are not in oe-core or poky?</div><div>(2:02:20 PM) khem: may be it needs more discussion</div><div>(2:02:38 PM) RP__: Jefro1: none atm as it was the poky bugzilla originally</div>
<div>(2:03:15 PM) Jefro1: thus no loss of information? </div><div>(2:03:26 PM) RP__: Jefro1: no</div><div>(2:03:27 PM) khem: ok should we carry it over to ml and ask devs about opinion</div><div>(2:03:33 PM) khem: ie. oe-core ml</div>
<div>(2:03:35 PM) RP__: khem: on what exactly?</div><div>(2:03:51 PM) RP__: khem: I'm trying to work out which question we're wondering about</div><div>(2:04:05 PM) khem: oecore category in yocto bugz</div><div>(2:04:14 PM) khem: I think its upto yocto too</div>
<div>(2:04:50 PM) RP__: Well, Yocto and OECore are going to remain pretty much in sync so should see the same bugs</div><div>(2:04:51 PM) khem: but should not yocto oe-core folks have a say that what I was thinking</div>
<div>
(2:05:11 PM) stefan_schmidt: I have to leave. We have all points of the agenda covered?</div><div>(2:05:19 PM) khem: ok</div><div>(2:05:21 PM) RP__: stefan_schmidt: yes, thanls!</div><div>(2:05:23 PM) khem: I think we are done</div>
<div>(2:05:34 PM) Jefro1: I'll send out minutes hopefully this afternoon</div><div>(2:05:36 PM) khem: the meeting is officially adjourned</div></div><div><br></div><div>-- </div><div>Jeff Osier-Mixon</div><font color="#888888"><div>
Yocto Project Community Manager @Intel</div>
<div><br></div>
</font></div><br><br clear="all"><br>-- <br>Jeff Osier-Mixon <a href="http://jefro.net/blog">http://jefro.net/blog</a><div>Yocto Project Community Manager @Intel <a href="http://yoctoproject.org">http://yoctoproject.org</a></div>
<br>