**** BEGIN LOGGING AT Thu May 29 02:59:56 2008
May 29 03:06:05 <cdbot2> * * OE Bug 4334 has been created by udovdh(AT)xs4all.nl
May 29 03:06:07 <cdbot2> * * dbus-native compilation fails
May 29 03:06:09 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=4334
May 29 03:06:19 <udovdh> need more info?
May 29 04:27:55 <CIA-22> 03khem 07org.oe.dev * r47696602... 10/ (4 files in 3 dirs): Add -frename-registers fix to GCC 4.3.0
May 29 04:27:59 <CIA-22> 03khem 07org.oe.dev * r318660b8... 10/ (3 files in 3 dirs): Prevent the loop in timespec_add_ns() to be optimised away
May 29 05:03:41 <jpieper> killall -9 vm1
May 29 05:05:24 <jpieper> oops, sorry
May 29 06:14:30 <Laibsch> good morning!
May 29 06:28:06 <cdbot2> * * OE Bug 4292 has been marked as DUPLICATE of bug 4212 by <Laibsch>
May 29 06:28:08 <cdbot2> * * Akita: no clicking on touchscreen possible
May 29 06:28:10 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=4292
May 29 06:30:30 <eFfeM> hi, i was wondering if there is support in oe for acpi on non intel platforms (e.g. mips)
May 29 06:30:50 <eFfeM> or does someone know about another place with info on this
May 29 06:31:04 <eFfeM> i am mainly interested in getting suspend-to-ram (S3) working on my device
May 29 06:35:46 <CoreDump> that is mostly a kernel thing if I am not mistaken.
May 29 06:36:11 <CoreDump> You can always build you kernel with an ACPI-enabled defconfig
May 29 07:17:50 <eFfeM> CoreDump, you are right, but the Kconfig for acpi says "depends on IA64 || X86"
May 29 07:18:08 <eFfeM> so clearly the mainstream kernel does not support mips or arm or so
May 29 07:18:29 <eFfeM> (this is the latest kernel from kernel.org)
May 29 07:19:58 <eFfeM> (but I actually guessed that in the handheld world suspend-to-ram would also be desirable, and hence I dropped the question here)
May 29 07:20:39 <XorA|gone> no arm device has ACPI AFAIK
May 29 07:21:02 <XorA|gone> suspend to ram already works for arm
May 29 07:21:10 <XorA|gone> and mips AFAIK
May 29 07:21:35 <eFfeM> i know but you do not even need acpi (you can tell the kernel to use a precompiled acpi table)
May 29 07:22:04 <eFfeM> XorA|gone, on what is it based then? do you have a pointer (either to a website or some file)
May 29 07:22:10 <XorA|gone> that makes no sense in arm world, a blank table is no different from no acpi
May 29 07:22:43 <XorA|gone> ACPI is a programming language which is pretty specific to x86 motherboards
May 29 07:23:19 <XorA|gone> eFfeM: arm emulates apm
May 29 07:23:38 <eFfeM> ah, ok, thought apm was being phased out in favour of acpi
May 29 07:23:57 <XorA|gone> in x86 world yes, but ACPI makes no sense for arm
May 29 07:24:10 <XorA|gone> AFAIK
May 29 07:24:15 <eFfeM> and it was my impression that acpi mainly did a mapping of functions to register access
May 29 07:24:27 <eFfeM> allowing hw abstraction
May 29 07:24:57 <XorA|gone> every arm pretty much has to define its own power model, all the SoCs are very different
May 29 07:25:04 <eFfeM> and the hw abstraction itself of course could be fairly platform independent
May 29 07:25:19 <eFfeM> hmm
May 29 07:25:31 <XorA|gone> dont forget on x86 all the IO is a different chips, most arms, IO is inside the cpu
May 29 07:25:39 <eFfeM> true
May 29 07:25:53 <XorA|gone> in fact some arms even have the ram inside the cpu
May 29 07:26:14 <eFfeM> actually my target is not a handheld, but a mips based embedded device, but it also has a lot of peripherals in the core
May 29 07:26:44 <XorA|gone> I suspect to suspend mips also uses apm emulation, I think some of the ipaqs are mips
May 29 07:27:13 <eFfeM> however i would still expect (both for arm and my mips device) that the i/o is still done using memory mapped registers etc
May 29 07:27:31 <eFfeM> of course there is probably no need for pci bus iteration and nasty things like that
May 29 07:28:34 <eFfeM> reading http://tuxmobil.org/apm_linux.html
May 29 07:28:40 <XorA|gone> but as the socs are hardwired rather than write some pseudo portable layer then write the binding, why not just write the binding :-)
May 29 07:29:32 <XorA|gone> anyway I have stuff todo and need to get moving
May 29 07:29:54 <eFfeM> agree, the only reason i can think of to use the pseudocode to create an interface layer and make things more standardised
May 29 07:30:07 <eFfeM> XorA|gone, thanks for your help & have a nice day!
May 29 07:48:07 <Laibsch> eFfeM: I am also interested in suspend
May 29 07:48:31 <Laibsch> In my case for the Zaurus, XorA|gone suggests it can already be done, but I never got around to really trying it out
May 29 07:48:43 <Laibsch> s/suspend/suspend to RAM/
May 29 07:48:53 <Genesis> bonjour
May 29 07:49:01 <Laibsch> to RAM, so that I can swap out an exhausted battery
May 29 07:49:31 <Laibsch> eFfeM: If there is anything remotely worth sharing, please let use know via the bug tracker
May 29 07:52:02 <Laibsch> eFfeM: I also have this among my bookmarks
May 29 07:52:05 <Laibsch> http://tree.celinuxforum.org/CelfPubWiki/SuspendToDiskForARM
May 29 07:52:48 <Laibsch> Of course, I meant suspend to DISK!, not suspend to RAM which while swapping out the battery would be pretty useless ;-)
May 29 08:24:22 <eFfeM> Laibsch, thanks for your reply, was afk to grab some coffee and ended up in another discussion :-)
May 29 08:25:09 <eFfeM> i'm aware of the celf pages, but never really got to digging into that part
May 29 08:25:41 <eFfeM> btw i vaguely recall someone else mentioned suspend to ram was also discussed in the latest celf conference
May 29 08:28:11 <frank7d> join #angstrom
May 29 08:46:39 <methril> morning
May 29 08:46:48 <methril> RP ping?
May 29 09:28:58 <likewise> HopsNBarley, HopsNBarley|work: http://www.pastebin.ca/1032949 - Whoot!
May 29 09:43:24 <RP> methril: pong
May 29 10:55:24 <methril> sorry, yesterday you mention some poky base.bbclass, are they located in the openembedded.dev repo?
May 29 10:55:59 <RP> methril: no, in pokylinux.org
May 29 10:56:09 <methril> thanks
May 29 11:31:04 <cdbot2> * * OE Bug 4326 has been RESOLVED (FIXED) by udovdh(AT)xs4all.nl
May 29 11:31:06 <cdbot2> * * gmp compilation fails
May 29 11:31:08 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=4326
May 29 11:35:52 <markus> i have built a custom made .bb-file providing virtual/kernel and placed it in my own bb-directory. when i tried to "bitbake virtual/kernel" bitbake threw a python stacktrace. Is this normal behaviour?
May 29 11:44:24 <RP> markus: It probably means there is a typo or something wrong with the .bb file
May 29 11:46:30 <Crofton|work> markus, for OSK?
May 29 11:49:40 <markus> Crofton|work: yup.
May 29 11:50:31 <markus> well... new custom made. just a copy of an existing one but places outside org.openembedded.dev
May 29 11:50:36 <markus> NOT custom made.
May 29 11:51:13 <markus> my goal is to use another defconfig
May 29 12:03:22 <Crofton> if you get a better defconfig, I'd be glad to push it
May 29 12:03:43 <Crofton> also, could you try building a 2.6.25 kernel? the patch is on muru.com now
May 29 12:04:26 <markus> okay... im an oe-newb but if i get anything to work ill inform you.
May 29 12:04:33 <Crofton> thanks
May 29 12:04:42 <Crofton> sorry I can't keep up with everything
May 29 12:04:55 <Crofton> having sound runnign again would be helpful for me also
May 29 12:05:07 <Crofton> I just have several other higher priority projects
May 29 12:18:05 <cdbot2> * * OE Bug 4335 has been created by likewise(AT)gmx.net
May 29 12:18:06 <cdbot2> * * netperf 2.4.4 does not build against eglibc
May 29 12:18:09 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=4335
May 29 13:12:40 <otavio> RP: what's the blockers for GIT tests? Are they only the issues at #4257?
May 29 13:13:08 <otavio> RP: I'd like to help 
May 29 13:13:23 <RP> otavio: The main blocker is me not having found time to sort some things out :(
May 29 13:13:29 <otavio> RP: I mailed ml yestarday (i guess it's held for approval)
May 29 13:13:51 <otavio> RP: couldn't I help to  sort them out?
May 29 13:14:16 <RP> otavio: By the time I explain what needs to be done I might as well have done it :/
May 29 13:14:47 <RP> otavio: Its a question of reaching an agreement amongst the core team really
May 29 13:14:49 * Laibsch cheers RP on to "go and have it done"
May 29 13:14:50 <Laibsch> ;-)
May 29 13:15:17 <Laibsch> RP: IOW, the tests are concluded already?
May 29 13:15:26 <RP> Laibsch: no
May 29 13:15:28 <Laibsch> core team sounds like decisions need to be made
May 29 13:15:31 <otavio> RP: I tend to agree but explaining it would allow others to help to choose the best solution available and avoid mistakes. Personally I like to have many opinions about sesrious issues
May 29 13:16:17 <RP> otavio: Its been discussed on the mailing list
May 29 13:16:51 <RP> its more a question of how the OE project can make a decision - who should make it and how?
May 29 13:17:11 <otavio> RP: Right but since I found this ticket and all related ones are rather trivial
May 29 13:17:23 <otavio> RP: except the one about mtn2git
May 29 13:17:29 <RP> otavio: Its not an accurate summary about the siutation
May 29 13:18:00 <otavio> RP: shouldn't then who is aware of the situation report the missing ones?
May 29 13:18:31 <RP> Some issues aren't easily represented in a bugzilla
May 29 13:18:54 <RP> Reporting them in the bugzilla isn't going to magically solve the problem or help in any way
May 29 13:18:57 <otavio> RP: there's someone else could give more details about what's missing?
May 29 13:19:16 <otavio> RP: I disagree. I could help in finding solutions
May 29 13:19:24 <otavio> RP: many others could too
May 29 13:19:54 <RP> otavio: Ok, so you're going to tell us how OE can make decisions?
May 29 13:20:18 <otavio> RP: that isn't a issue but a team agreement. That I can't help.
May 29 13:20:23 <otavio> RP: i  mean technical ones
May 29 13:20:32 <RP> otavio: The problems are not technical in nature
May 29 13:21:10 <otavio> RP: but using a voting system seems to work well in a project of size of Debian. OE seems smaller
May 29 13:21:17 <Laibsch> RP: So, how are we going forward, then?
May 29 13:21:37 <RP> Laibsch: I am talking to people and I've made agreements on a way forward
May 29 13:21:46 <RP> The person who I hoped would help with that doesn't have time
May 29 13:21:46 <Laibsch> OK
May 29 13:21:57 <RP> I've made arrangements to try and do that work myself now
May 29 13:22:09 <RP> I just don't have the time to actually do it :(
May 29 13:22:35 <Laibsch> Basically, the test stuff has become secondary and we are back to "personal and politics" (not meant in a negative way, trying to understand)
May 29 13:22:45 <RP> I have a ton of issues coming from many different directions and I'm trying to prioritise them as best I can :(
May 29 13:22:55 <RP> Laibsch: personal and politics was always the primary problem
May 29 13:22:56 <Laibsch> Understood
May 29 13:23:01 <Laibsch> I know
May 29 13:23:20 <Laibsch> I thought that the test server might help in "convincing" 
May 29 13:23:26 <Laibsch> One side or the other, even
May 29 13:23:34 <Laibsch> in case anything unexpected showed up
May 29 13:23:39 <RP> The test server is part of the discussion
May 29 13:23:50 <Laibsch> Anything left to do there
May 29 13:23:51 * otavio can't understand ... looks like the most active "partners" of OE use GIT ... what others politics could exist?
May 29 13:23:53 <Laibsch> ?
May 29 13:24:10 <RP> Laibsch: commit hooks would be good to get setup
May 29 13:24:18 <Laibsch> otavio: A major contributor dislikes git with a passion
May 29 13:24:26 <Laibsch> contributor to OE
May 29 13:24:31 <RP> Laibsch: more than one
May 29 13:24:39 <otavio> Laibsch: I guess this should be koen (from the ml thread)
May 29 13:25:04 <Laibsch> now that RP mentioned it, I can remember another contributor
May 29 13:25:11 <Laibsch> And the number is not the issue
May 29 13:25:44 <Laibsch> The fact is that other than debian with a large number, even a single person lost can mean big trouble for OE
May 29 13:26:05 <Laibsch> not that I think that OE should let itself be held ransom because of that, though
May 29 13:26:14 <otavio> Laibsch: we all have problems when important people dislikes something in project
May 29 13:26:20 <otavio> Laibsch: we have it in Debian all the time
May 29 13:26:25 <Laibsch> And I am beginning to see a bit of that (because of the lack of *open* discussion)
May 29 13:26:36 <otavio> Laibsch: but usually this doesn't hold the project to move forward
May 29 13:26:51 <Crofton> we've had the discussion on -dev several times
May 29 13:26:53 <RP> Almost all of the discussion has been open
May 29 13:27:07 <otavio> RP: hehe "almost" is funny
May 29 13:27:12 <Laibsch> Well, there hasn't been much news for several weeks
May 29 13:27:19 <RP> open doesn't magically make it good/right/better
May 29 13:27:28 <Crofton> there hasn't been any discussion :)
May 29 13:27:29 <otavio> RP: sure. I agree
May 29 13:27:45 <RP> Laibsch: There has been no change for several weeks
May 29 13:27:47 <Laibsch> And last thing I remember is that people agreed on "let's test things out and make a decision then"
May 29 13:27:58 <Laibsch> that does not seem to be the full truth anymore
May 29 13:28:02 <RP> Laibsch: That is still the position
May 29 13:28:04 <Crofton> ?
May 29 13:28:15 <otavio> RP: wouldn't be better to have a deadline?
May 29 13:28:15 <Crofton> hasn't seen anything to the contrary
May 29 13:28:37 <RP> otavio: No and I have nothing more to say on the subject
May 29 13:28:41 <otavio> RP: FOSS projects tend to work better with deadlines hehe
May 29 13:28:45 <Crofton> heh
May 29 13:28:49 <Laibsch> RP: If that is still the position, we should welcome otavio's offer for getting the remaining open bugs fixed and then document what is good and bad for OE wrt to git
May 29 13:29:04 <Crofton> otavio, a bunch of guys are at linuxtag atm
May 29 13:29:25 <Crofton> and there should be an "official" OE organization soon
May 29 13:29:31 <RP> Laibsch: Whats needed is for someone on the core team who understands the politics to try and move this forward
May 29 13:29:52 <Crofton|work> and typing on the laptop kills my arm
May 29 13:30:03 <RP> Some are too quiet to do it, some don't have time etc
May 29 13:30:06 <otavio> Crofton: I can imagine ...
May 29 13:30:06 * Laibsch is not even sure who is a member of that "secret society" ;-)
May 29 13:30:10 <Laibsch> or semi-secret
May 29 13:30:18 <Crofton|work> Laibsch, rubbish :)
May 29 13:30:18 <RP> Laibsch: Its listed on the website
May 29 13:30:59 <Laibsch> Crofton: I was talking a bit tongue in cheek.  And I was told it's on the website before, but IIRC there is at least one person missing
May 29 13:31:02 * Laibsch looks
May 29 13:31:42 <RP> For better or for worse I'm the member of the core team pushing the SCM issue atm. If a suitable replacement can be found I'd welcome the oppertunity to drop it from my todo list since I hate politics...
May 29 13:31:45 <Laibsch> nope, I was mistaken
May 29 13:32:09 <Laibsch> mickeyl is *not* core?
May 29 13:32:28 <otavio> RP: I can't help in politics but please count on me about technical stuff
May 29 13:32:49 <Laibsch> thanks, otavio
May 29 13:32:59 <Laibsch> RP: mickeyl is *not* core?
May 29 13:33:06 <otavio> Laibsch: you and OE are welcome ;-)
May 29 13:33:11 <RP> Laibsch: he is
May 29 13:33:19 <Laibsch> well, then I was correct
May 29 13:33:28 <RP> otavio: Thanks for the offer it is appreciated and if you can help I will let you know
May 29 13:33:33 <Laibsch> the website is not reflecting things correctly
May 29 13:33:45 <RP> Laibsch: As a founder he has a kind of special position
May 29 13:33:50 <Laibsch> And even the transfer of core = "core + founders" is probably not correct
May 29 13:34:05 <RP> Laibsch: Some of the founders are inactive, yes
May 29 13:34:26 <Laibsch> so, it is "those listed as core on http://oe.linuxtogo.org/people + Mickey"?
May 29 13:34:28 <otavio> There're someone here that works with Digital TV?
May 29 13:34:39 <likewise> otavio: somewhat
May 29 13:34:51 <likewise> otavio: depends on what you mean with digital tv
May 29 13:34:51 <otavio> likewise: you do?
May 29 13:34:51 <RP> Laibsch: mickey is certainly considered part of the core team
May 29 13:35:13 <otavio> likewise: the framework/applications to be used on digital tv systems
May 29 13:35:28 <otavio> likewise: or embedded OS for it
May 29 13:35:32 <otavio> likewise: anything related
May 29 13:35:40 <Laibsch> I guess, kergoth and schurig are not since they have been inactive?  IOW, there are six members overall?
May 29 13:35:43 <likewise> otavio: depends on what you mean with digital tv. I'm mainly in the 3 Gigabit/seconds domain for digital transmission of TV
May 29 13:36:35 <otavio> likewise: http://en.wikipedia.org/wiki/Digital_tv
May 29 13:37:02 <RP> Laibsch: as I understand it
May 29 13:37:11 <likewise> otavio: www.axon.tv
May 29 13:37:12 <otavio> likewise: we're beggining to use that here in brazil and i'd like to talk to someone who works with that
May 29 13:37:21 <RP> Laibsch: Once we get a foundation setup its likely we'll rework things somehow
May 29 13:37:30 <Laibsch> OK
May 29 13:37:35 <Laibsch> Thanks for your time
May 29 13:37:40 <likewise> otavio: just ask
May 29 13:37:55 * Laibsch hopes that the imminent e.V. will improve steering for OE
May 29 13:38:03 <otavio> likewise: i guess it's off-topic .. mind to talk privately?
May 29 13:38:20 <RP> Laibsch: I hope so too, we've kind of waiting on that rather than trying to change things now
May 29 13:38:20 <chouimat|work> Laibsch: or make it worse ... 
May 29 13:38:53 <otavio> RP: thanks by making things clear
May 29 13:40:02 <Laibsch> RP: And the waiting was sometimes unbearable (at least for me), because it stalled/stalls many things
May 29 13:40:08 <Laibsch> chouimat|work: I hope not
May 29 13:40:20 <Laibsch> At least we can make decisions with some formality
May 29 13:40:35 * RP has learned patience
May 29 13:40:53 <RP> most things do work out in time. Some just take longer than others :)
May 29 13:40:56 <Laibsch> And we can't put off decisions by pointing to "we will have an e.V. real-soon-now[TM]
May 29 13:41:32 <RP> I think I've made it clear that is not the case
May 29 13:41:33 <Laibsch> patience is a virtue, so is impatience ;-)
May 29 13:43:09 <chouimat|work> Laibsch: most projects I know that got a e.V. become political black holes ... aka they suck all the fun :(
May 29 13:43:23 <CIA-22> 03john_lee 07org.oe.dev * r0de0caea... 10/ (1 packages/dbus-java/dbus-java_2.4.bb): dbus-java: add 2.4
May 29 13:43:27 <CIA-22> 03john_lee 07org.oe.dev * r35eca457... 10/ (4 files in 3 dirs): libmatthew: add 0.7
May 29 13:43:41 <udovdh> hello
May 29 13:43:56 <Laibsch> RP: I am not accusing anyone.  But decisions are slow.  And this hurts other things depending on it which are then also postponed.  
May 29 13:43:56 <Laibsch> Just a stupid and simple example is the openembedded-common stuff I did (the initial setup with mtn took me some time to figure out and I dropped the ball a few times).  I have been thinking about this for six months to a year, I think.
May 29 13:44:31 <Laibsch> chouimat|work: Then you'll be our primary consultant on how to avoid those mistakes 
May 29 13:45:32 <chouimat|work> Laibsch: hehe
May 29 13:47:31 * Laibsch is not even listed as dev on http://oe.linuxtogo.org/people
May 29 13:48:09 <udovdh> can anybody here help me with this compile error? http://bugs.openembedded.net/show_bug.cgi?id=4334
May 29 13:48:48 <udovdh> I am progressing in my attempt to build navit with OE (angstrom-2007.12), closed another bug today...
May 29 13:48:49 * Crofton isn't either
May 29 13:48:59 <Crofton> but, that is my fault
May 29 13:49:01 * udovdh ?
May 29 13:49:15 <Laibsch> udovdh: Maybe you can try to look if there is any significant difference between .stable and .dev?
May 29 13:49:27 <udovdh> I am using stable.
May 29 13:49:34 <Laibsch> I know
May 29 13:49:41 <udovdh> stable for all of it?
May 29 13:49:46 <Laibsch> Have you tried compiling against .dev?
May 29 13:49:47 <udovdh> or just for dbus?
May 29 13:49:58 <udovdh> earlier yes but also found (other) bugs
May 29 13:49:58 <Laibsch> If it compiles there, then you should inspect the differences
May 29 13:50:03 <Laibsch> sure
May 29 13:50:04 <udovdh> hmmmz
May 29 13:50:11 <udovdh> ok, will try
May 29 13:50:40 <Laibsch> "xxdiff org.openembedded.dev/packages/dbus org.openembedded.stable/packages/dbus"
May 29 13:50:42 <Laibsch> works wonders
May 29 13:51:12 <markus> Crofton|work: using the defconfig for kernel 2.6.24 provided by openembedded when compiling 2.6.25 works, but when using defconfig copied from linux.2.6.25somethingsomething.tar.gz/arch/arm../configs/omap_osk_5912_defconfig from kernel.org fails.
May 29 13:51:21 <markus> that is old information or what?
May 29 13:51:28 <udovdh> danke schoen Laibsch 
May 29 13:51:40 <Crofton> fails how?
May 29 13:51:40 <Laibsch> Crofton: You are listed, aren't you?  In Core.
May 29 13:51:50 <Laibsch> udovdh: Keine Ursache
May 29 13:52:12 <markus> i haven't saved the output but it complains about something nokia770
May 29 13:52:14 <woglinde> hi
May 29 13:52:38 <Crofton> ah compile failuer
May 29 13:52:43 <markus> yes.
May 29 13:53:08 <Crofton> Laibsch, no, Philip Balister
May 29 13:53:42 <Laibsch> ah, pb would have matched.  I thought you were the one.
May 29 13:53:57 <Laibsch> no other pb around here, either
May 29 13:53:58 <Crofton> heh no
May 29 13:54:02 <Laibsch> ~seen pb
May 29 13:54:04 <ibot> pb <n=pb@2002:5246:d929:1:20e:2eff:fe2d:60bf> was last seen on IRC in channel #gpe, 898d 4h 30m 28s ago, saying: 'florian_hrz: good morning!'.
May 29 13:54:15 <Laibsch> ~seen pb_
May 29 13:54:15 <Crofton> ~seen pb_
May 29 13:54:16 <ibot> pb_ <n=pb@castle.reciva.com> was last seen on IRC in channel #oe, 45d 3h 55m 9s ago, saying: 'zecke, florian: good morning'.
May 29 13:55:03 <Crofton> markus, I'm afraid not many people are testing the OSK from git :(
May 29 13:55:25 <udovdh> I set stuff up using the info at http://www.angstrom-distribution.org/building-angstrom
May 29 13:55:34 <udovdh> so how do I switch to dev?
May 29 13:55:41 <udovdh> it's not just changing the BBPATH etc
May 29 13:55:52 <Crofton> marcan, maybe loo in the linux-omap list and see if there is a config trick you can do to get it to build
May 29 13:56:20 <Crofton> I need to close this window and write an abstract for a talk next week
May 29 13:56:58 <udovdh> aha
May 29 13:57:01 <udovdh> also gotta change distro
May 29 13:57:41 <udovdh> to angstrom-2008.1
May 29 14:02:31 <lumon> oe
May 29 14:07:17 <lumon> oe
May 29 14:07:34 <likewise> oe yeah!
May 29 14:11:04 <lumon> why
May 29 14:20:33 <likewise> lumon: why what?
May 29 14:26:46 <HopsNBarley> likewise, just saw your pastebin - awesome!
May 29 14:27:47 <likewise> HopsNBarley: yup, nice processor.
May 29 14:28:15 <likewise> HopsNBarley: I wonder if Marvell has an ARM competitor. This is the lowest end PowerPC and I need to scale up.
May 29 14:28:46 <likewise> HopsNBarley: it can reach 800 Mbit/s UDP receive, and 300 Mbit/s under a real-time kernel, I just finished some testing.
May 29 14:29:08 <HopsNBarley> that is pretty awesome for a SOC.
May 29 14:29:15 <likewise> HopsNBarley: throughput should be better for non-real-time
May 29 14:29:28 <likewise> HopsNBarley: I'm now testing the SATA interface (1.5 / 3.0 Gbps)
May 29 14:29:58 <HopsNBarley> back of the envelope, i'd guess the marvel orion would be comparable. 
May 29 14:37:58 <woglinde> hm tomorrow is my linuxtag turn
May 29 14:41:11 <likewise> woglinde: talk?!
May 29 14:41:32 <woglinde> nope
May 29 14:41:35 <woglinde> on the booth
May 29 14:41:38 <markus> Crofton: have you had sound working on the omap5912osk, and what kernel version?
May 29 15:06:57 <Crofton|work> markus, I forget, it was a wile backl :(
May 29 15:09:41 <markus> Crofton|work: 2.6 kernel?
May 29 15:12:30 <Khem> HopsNBarley: hey
May 29 15:12:54 <HopsNBarley> hi Khem - just headin' to work.
May 29 15:13:09 <Khem> HopsNBarley: cool see you laters
May 29 15:13:10 <woglinde> hi khem
May 29 15:13:15 <Khem> woglinde: hey
May 29 15:13:22 <HopsNBarley> excellent - i'll be on at work...
May 29 15:13:27 <Crofton|work> markus, yes
May 29 15:13:38 <Khem> HopsNBarley: so we will meet at 11:45 in front of Blockbuster today 
May 29 15:14:13 <HopsNBarley> sounds great!
May 29 15:15:01 <likewise> hi Khem, say hi to HopsNBarley from me :-)
May 29 15:15:18 <likewise> khem: and have a nice lunch as well
May 29 15:15:34 <Khem> I am seeing an issue with gcc 4.3 where it is not finding C++ headers from staging where as gcc 4.2.2 is built same way packaged same way it finds them
May 29 15:15:39 <Khem> likewise: will do :)
May 29 15:16:11 <Khem> if I copy back the c++ headers to original location in cross/ then it finds them
May 29 15:16:38 <Khem> --gxx-include-dir is not being honored
May 29 15:16:57 <likewise> khem: hmmn, is this a known bug upstream?
May 29 15:17:13 <Khem> likewise: I have hit it today
May 29 15:17:26 <Khem> I did not search yet
May 29 15:17:59 <Khem> I wanted to find out if someone else is running into same issue as me or not
May 29 15:18:40 <Khem> Laibsch was trying gcc 4.3 I dont know if he hit this issue or not
May 29 15:20:03 <Laibsch> rwhitby: Are you awake and online?
May 29 15:20:13 <Laibsch> Khem: I don't think so
May 29 15:20:19 <Laibsch> no gcc issues lately
May 29 15:20:59 <Khem> Laibsch: do u have gcc-4.3 built oe handy ?
May 29 15:21:13 <Khem> Laibsch: can you try to bitbake libusb
May 29 15:22:02 <Khem> I am also seeing --oldinclude --includedir being specifies for cross-gcc I dont see reason
May 29 15:22:08 <Laibsch> Khem: "bitbake gcc-native-4.3 && bitbake libusb"?
May 29 15:22:52 <Khem> Laibsch: I mean you must have gcc 4.3 as you default cross gcc and then build libusb using this cross-gcc
May 29 15:23:24 <Laibsch> sorry, too much trouble right now
May 29 15:23:42 <Laibsch> If gcc 4.3 is default I can give it a try
May 29 15:23:45 <Khem> We move the c++ headers into staging and tell gcc to use them from there using --gxx-include-dir which is all we have to do
May 29 15:23:49 <Laibsch> for example default in A* 2008
May 29 15:23:52 <Khem> Laibsch: yes
May 29 15:23:59 <Laibsch> OK, let me see
May 29 15:24:12 <Khem> Laibsch: I think 4.2.2 is default from a*
May 29 15:24:26 <Khem> Laibsch: you might have to edit it in conf file
May 29 15:24:31 <otavio> a*?
May 29 15:24:39 <Khem> angstrom :)
May 29 15:24:43 <otavio> ahhhhhhhhh
May 29 15:24:54 <otavio> ah+ ;-)
May 29 15:25:29 * Khem otavio should have used while loop 
May 29 15:26:29 <Laibsch> Khem: It's not default
May 29 15:26:37 <Laibsch> "bitbake gcc-cross" builds 4.2
May 29 15:26:40 <Laibsch> sorry, you're out
May 29 15:26:47 <Laibsch> Can't help you
May 29 15:30:04 <Khem> Laibsch: ok
May 29 15:30:19 <Khem> Laibsch: if you try out a build then try 4.3 :)
May 29 15:30:29 <Khem> I have fixed few stuff in it
May 29 15:31:24 <Khem> i am seeing that includedir and oldincludedir is not needed after sysroot stuff I believe
May 29 15:31:31 <Khem> I will experiment with it
May 29 15:33:11 <CIA-22> 03koen 07org.oe.dev * rc56cf748... 10/ (1 conf/distro/include/angstrom-2008-preferred-versions.inc): angstrom: prefer cairo 1.6.4
May 29 15:33:15 <CIA-22> 03koen 07org.oe.dev * r2c36c135... 10/ (1 packages/abiword/abiword_2.6.3.bb): abiword: add 2.6.3
May 29 15:38:04 <cdbot2> * * OE Bug  has been RESOLVED by jeremy.laine(AT)bolloretelecom.eu
May 29 15:38:06 <cdbot2> * * oestasts: URL for searches
May 29 15:38:08 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=
May 29 15:40:19 <Crofton|work> hmm, glibc-intermediate just failed for me
May 29 15:40:29 <Crofton|work> Shoudl I take my own advice and nuke tmp
May 29 15:41:31 <woglinde> yes
May 29 15:41:32 <woglinde> *g*
May 29 15:41:38 <Crofton|work> :)
May 29 15:41:52 <woglinde> but I didnt compile glibc for ages
May 29 15:41:56 <woglinde> uclibc works fine
May 29 15:42:03 <woglinde> and compiles faster
May 29 15:46:14 <Khem> Crofton|work: use eglibc :)
May 29 15:46:21 <Khem> Crofton|work: it works
May 29 15:46:49 <woglinde> hehe
May 29 15:46:54 <woglinde> re schurig
May 29 15:47:25 <Khem> Hmmm I cant find where oldincludedir is specified
May 29 15:47:39 <woglinde> hm
May 29 15:47:42 <woglinde> ask RP
May 29 15:48:05 <Khem> strange find is failing to find it in package gcc yet it appears in config when built
May 29 15:48:31 <RP> Khem: oldincludedir comes from autotools.bbclass
May 29 15:48:40 <Khem> RP: cool 
May 29 15:48:44 <woglinde> hehe
May 29 15:48:46 <woglinde> hi RP
May 29 15:48:52 <RP> hi woglinde 
May 29 15:48:57 <Khem> RP: I was about to do a full search
May 29 15:49:26 <Khem> RP: --includedir and --oldincludedir should be pointing into staging 
May 29 15:49:35 <Khem> right now they point into cross
May 29 15:50:01 <RP> Khem: ah, probably a bug with cross.bbclass
May 29 15:50:15 <Khem> RP: OK
May 29 15:51:05 <cdbot2> * * OE Bug  has been RESOLVED by jeremy.laine(AT)bolloretelecom.eu
May 29 15:51:07 <cdbot2> * * oestast: unexpand package view by default
May 29 15:51:09 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=
May 29 15:51:36 <jeremy_laine> Laibsch: ping?
May 29 15:51:57 <Khem> oldincludedir = "${exec_prefix}/include"
May 29 15:52:19 <Khem> RP: sso this needs to be variable pointing into staging
May 29 15:53:08 <Khem> RP: btw. Do we really need these two now that we have sysroot ?
May 29 15:53:31 <RP> Khem: It will default to /usr/include which is really wrong
May 29 15:53:45 <Khem> RP: ah for other packages
May 29 15:54:06 <RP> Khem: For other packages it will just be /usr/include
May 29 15:54:06 <Khem> not for gcc though we use sysroot there so sysroot will be appended
May 29 15:54:22 <Khem> RP: got it
May 29 15:54:41 <RP> Khem: We have that since we do support changing the layout, at least in priciple
May 29 15:54:53 <Khem> ok 
May 29 15:56:01 <Laibsch> jeremy_laine: I'm here
May 29 15:56:05 <Laibsch> go right ahead
May 29 15:56:12 <Khem> RP: can I use  ${STAGING_INCDIR} here in cross.bbclass
May 29 15:56:14 <Laibsch> thanks for the updates to oestats
May 29 15:56:38 <jeremy_laine> Laibsch: if you want to give the new version a spin take a look at my devel server (http://oestats.jerryweb.org/)
May 29 15:56:46 <RP> Khem: maybe not, I think that will point at the wrong place
May 29 15:56:47 <jeremy_laine> Laibsch: I have pinged koen to do an update on amethyst
May 29 15:56:51 <dcordes> what is the preferred way to make your kernel recipe build a defconfig which is inside the desired kernel tree?
May 29 15:57:07 <Khem> RP: hmmm why
May 29 15:57:13 <RP> Khem: ${STAGING_DIR_TARGET}/${layout_includedir} should work
May 29 15:57:19 <Khem> RP: oh yes
May 29 15:57:39 <Khem> RP: Do we depend on any headers from cross at all?
May 29 15:57:42 <RP> Khem: STAGING_INCDIR will point at ${STAGING_DIR_HOST}${layout_includedir}
May 29 15:57:54 <RP> Khem: Only the internal gcc ones
May 29 15:57:57 <Khem> RP: ok saw it
May 29 15:57:59 <Laibsch> Khem: libusb built just fine after gcc-cross 4.3.0, here
May 29 15:58:01 <Laibsch> http://tinderbox.openembedded.net/builders/Laibsch/
May 29 15:58:27 <Khem> Laibsch: ok
May 29 15:58:55 <Laibsch> jeremy_laine: Thanks.  Looks very nice
May 29 16:01:13 <Khem> Laibsch: I hope it used gcc 4.3 
May 29 16:02:17 <dcordes> I would like to update the linux-kaiser.bb to build from git.linuxtogo.org rather than from git.android.com. In its current state it uses a patch that applies a .config directly usable. In git.linuxtogo.org I added a htckaiser_defconfig. Now I wonder how I can move it to .config. Is there any mechanism for this?
May 29 16:02:46 <Khem> time to go to offce bye all
May 29 16:29:15 <thesing> hi everybody
May 29 17:41:23 <woglinde> re
May 29 17:41:25 <woglinde> hi thesing
May 29 18:12:58 <Crofton> Linuxtaggers: http://beagleboard.blogspot.com/2008/05/beagle-board-at-linuxtag.html
May 29 18:14:02 <woglinde> hehe
May 29 18:14:06 <woglinde> Klick
May 29 18:17:43 <CosmicPenguin> I am told that the OE booth is near the coreboot booth - so stop by
May 29 18:17:43 <CosmicPenguin> mmmm.... free x86 firmware....
May 29 18:19:05 <woglinde> hm
May 29 18:19:19 <woglinde> I will at the oe booth tomorrow
May 29 18:19:23 <woglinde> +be
May 29 18:20:00 <Crofton> woglinde, go say high to the beagle guys
May 29 18:20:14 <Crofton> and make sure they are running code built with OE :)
May 29 18:20:53 <woglinde> hahahaa
May 29 18:21:35 <woglinde> I will show my simpad running whith uclibc, angstroem and opie
May 29 18:22:08 <Crofton> they should have Angstrom running
May 29 18:22:21 <Crofton> I think we've made the most complete demo image for them
May 29 18:22:24 <woglinde> hm I meant at our booth
May 29 18:22:30 <woglinde> not to the beagle guys
May 29 18:23:12 <thesing> you'll have a hard time to fine enough place on the table ;)
May 29 18:23:32 <thesing> it's already crowded with devices.
May 29 18:23:58 <Crofton> woglinde, just go to their booth and say hi
May 29 18:24:33 <woglinde> ah thesing
May 29 18:24:45 <woglinde> will you be theire tomorrow too?
May 29 18:24:57 <woglinde> I take it with me and we will ee
May 29 18:25:01 <woglinde> aeh see
May 29 18:25:33 <thesing> unfortunately not. I was there yesterday and today. But I have to work at least some days a week.
May 29 18:25:45 <woglinde> ;)
May 29 18:57:55 <CIA-22> 03crofton 07org.oe.dev * ra9d44fae... 10/ (4 files in 3 dirs): 
May 29 18:57:55 <CIA-22> linux-omap1 : Add bb file to build 2.6.25 kernel.
May 29 18:57:55 <CIA-22>  Build tested only.
May 29 18:59:08 <Crofton> markus, ping
May 29 19:06:03 <markus> Crofton: pong pong
May 29 19:06:43 <Crofton> markus, I just commited files that build .25 with sound on
May 29 19:06:51 <Crofton> I think :)
May 29 19:06:53 <markus> oh... i built a 2.6.25 kernel today, commented out something about nokia 770. but i could not load the alsa modules
May 29 19:07:05 <Crofton> I built sound into the kernel
May 29 19:07:35 <Crofton> do you mind taking a look at what I commited and filing bug reports?
May 29 19:07:37 <markus> ooh, not a module? you know i'm not an expert on there things. i will take a look on it
May 29 19:08:19 <Crofton> as long as the kernel fits in the partition, I don't see any reason to modularize the kernel for peripherals that are on the board
May 29 19:09:00 <markus> i can't test it today. but tomorrow
May 29 19:09:05 <Crofton> thanks
May 29 19:09:40 <markus> no thank you
May 29 19:10:18 <markus> but i failed to load some kind of alsa-aic23 module today, and the system froze
May 29 19:11:10 <markus> http://endata.olf.sgsnet.se:8080/build/tmp/deploy/glibc/ipk/omap5912osk/  <--- if you want to take a look
May 29 19:11:39 <markus> maybe not, ill try it tomorrow
May 29 19:22:42 <Fidget> I do not understand how to include support in a custom image for a specific RRECOMMENDS_task-boot-mysystem.
May 29 19:23:09 <Fidget> Essentially I want a refined version of: BUILD_ALL_DEPS = "1", is there anything like that?
May 29 19:27:56 <Fidget> Can anyone assist with helping me understand how to build RRECOMMENDS?
May 29 19:30:21 <thesing> Fidget: what is your problem with it?
May 29 19:31:20 <thesing> I'm not sure what you are trying to do.
May 29 19:33:10 <woglinde> re
May 29 19:35:43 <woglinde> fidget all what is in IMAGE_INSTALL = " " is installed in an image
May 29 19:45:12 <Fidget> woglinde: is it reasonable for me to put IMAGE_INSTALL = "tasks-base-list RRECOMMENDS_tasks-base-list-subset"
May 29 19:49:49 <thesing> Fidget: no
May 29 19:50:24 <Fidget> thesing: Do you have a better idea of what I am attempting to accomplish?
May 29 19:50:41 <thesing> Fidget: you would create a task-base-subset and replace task-base-list with task-base-subset
May 29 19:51:10 <Fidget> thesing: so copy the task-base-list to task-base-subset and change the RRECOMMEND to an RDEPENDS?
May 29 19:51:45 <thesing> yes this is a way.
May 29 19:51:56 <Fidget> is there another way?
May 29 19:52:19 <thesing> I'm not sure if IMAGE_INSTALL pulls RRECOMMEND by default.
May 29 19:52:55 <Fidget> BUILD_ALL_DEPS seems to toggle that according to: http://www.openembedded.info/usermanual.html#image_class
May 29 19:53:03 <thesing> What is the problem you want do solve?
May 29 19:53:50 <Fidget> Rather than building all RRECOMMENDS, I want to tell the custom image to only build RRECOMMENDS_tasks-base-list-subset
May 29 19:53:58 <Fidget> err, s/build/include/
May 29 19:54:57 <thesing> so you want to exclude some packages from the image that are RRECOMMENED?
May 29 19:55:31 <Fidget> Well, I'd like to include none of the RRECOMMEND and then add only the ones I want
May 29 19:57:27 <thesing> wait a moment.
May 29 19:59:19 <thesing> its seems that there is no such way currently but it should be easy to add.
May 29 19:59:26 <Fidget> ok
May 29 20:00:19 <thesing> opkg-cl decides what it pulls into images.
May 29 20:01:07 <thesing> there should be an command line option to tell it that you don't want to pull RRECOMMENDS
May 29 20:02:31 <thesing> you can then expose this as PACKAGE_INSTALL_NO_DEPS does. (see classes/rootfs_ipk.bbclass)
May 29 20:03:03 <Fidget> thesing: I'm going with the tasks/tasks-subset route
May 29 20:04:59 <thesing> after I looked into the code I'm not sure if this will work. I think opkg-cl always pull the rrecommends. Of course you could add set PACKAGE_INSTALL_NO_DEPS to your image but that you won't even get RDEPENDS.
May 29 20:05:21 <Fidget> nah
May 29 20:06:46 <thesing> but again I'm not sure if opkg-cl pull rrecommends. But if it wouldn't you wouldn't ask I guess.
May 29 20:06:49 <Fidget> thesing: Thanks a ton for your input
May 29 20:10:25 <thesing> you're welcome.
May 29 20:12:37 <Fidget> thesing: unfortunately that means you've been moved up my totem pole of useful sources of OE information....
May 29 20:13:00 <thesing> it seems that there is no option to teach opkg-cl to don't pull rrecommends. 
May 29 20:13:35 <thesing> that's no problem.
May 29 20:57:37 <cyril__> T2 and OpenEmbedded "compared": http://www.t2-project.org/handbook/html/t2.intro.different.distribution.html
May 29 20:57:43 <cyril__> .. ridiculous
May 29 20:58:16 <woglinde> cyril aeh yes there are some ugh "otheminded" people in the t2 project
May 29 20:58:26 <woglinde> "otherminded"
May 29 20:58:41 <woglinde> I didnt want to use psychopatic
May 29 20:58:58 <cyril__> "otherminded" ?
May 29 20:59:19 <cyril__> I'm not against people that think different (I'm even for that!)
May 29 20:59:32 <cyril__> but starting with "proprietary BitBake based"
May 29 21:00:00 <cyril__> and the rest is only '???', i.e. no study at all
May 29 21:00:08 <cyril__> this comparison is a joke
May 29 21:00:51 <cyril__> it's as if I would say T2 is crap, whereas I don't know if it's or not
May 29 21:01:21 <woglinde> cyril the one who wrote this joinded the channel and wasnt very nice
May 29 21:01:47 <cyril__> woglinde: too bad :-(
May 29 21:01:49 <woglinde> he is one of the type joerg schilling
May 29 21:01:58 <woglinde> if the name says something to you
May 29 21:01:59 <cyril__> woglinde: a fair and scientific comparison would be nice
May 29 21:04:22 <cyril__> woglinde: I don't know him, but 4th result from google is 'Jorg Schilling is damage', which is not a good point for him :-/
May 29 21:04:34 <Crofton|work> that article is insna
May 29 21:04:35 <Crofton|work> e
May 29 21:11:37 <kgilmer> it seems like the T2 site uses the same color scheme as OE
May 29 21:11:53 <kgilmer> good thing colors aren't proprietary.
May 29 21:17:56 <szsoftware> Hello everybody!
May 29 21:28:14 <szsoftware> Is someone able helping me with a OE problem? It is about fetch failed, when fetching from git. Thanx in advance!
May 29 21:28:33 <Crofton|work> can you pastebin the message
May 29 21:29:26 <woglinde> szsoftware try to remove the git dir in your download-dir and try again
May 29 21:29:37 <thesing> bye
May 29 21:29:48 <woglinde> bite thesing
May 29 21:29:50 <woglinde> aeh nite
May 29 21:30:09 <thesing> *lol*
May 29 21:30:29 <thesing> I'm not that tasty I think ;)
May 29 21:30:43 <szsoftware> I will have a try following suggestion in #htc-linux. Thanx
May 29 21:34:36 <kgilmer> I asked this in the poky channel but didn't get a response.  Is there a way to prepend a source stash?
May 29 21:34:57 <kgilmer> I'd like to speed up my builds by specifying an internal server that has all source tarballs .
May 29 21:36:13 <kgilmer> I tried defining SRC_TARBALL_STASH in my local conf
May 29 21:36:25 <Fidget> wtf is autodetected dependencies for T2?
May 29 21:36:29 <Fidget> that's insane
May 29 21:56:15 <szsoftware> Crofton|work: http://pastebin.com/m6e60f991
May 29 21:56:43 <Crofton|work> you need to specify source rev, let me fin an exampel
May 29 21:57:08 <Crofton|work> SRCREV = "6b403428164c53d94bd9377d84e69052563dd57d"
May 29 21:57:15 <Crofton|work> add that to the bb file
May 29 21:57:28 <Crofton|work> well, use the string for the git rev you want :)
May 29 21:59:27 <szsoftware> Ok, thanks, just tried another suggestion, that doesn't work
May 29 22:00:14 <Crofton|work> look in packages/linux/linux-omap2_git.bb 
May 29 22:00:31 <ant_> good morning #OE
May 29 22:00:52 <szsoftware> ant_: good morning
May 29 22:01:06 <ant_> RP: ping (regression found)
May 29 22:01:19 <ant_> RP: it's udev
May 29 22:01:36 <ant_> szsoftware: hey
May 29 22:02:53 <woglinde> hi ant
May 29 22:04:33 <ant_> woglinde: hey, wie geht es in Berlin?
May 29 22:06:11 <woglinde> ant good
May 29 22:06:23 <woglinde> today will be linuxtag
May 29 22:07:15 <ant_> sorry, too busy to visit you...
May 29 22:07:18 <szsoftware> Oh really?
May 29 22:07:27 <szsoftware> Is'nt it in Berlin?
May 29 22:07:28 <Crofton> szsoftware, my suggestion should solve your problem
May 29 22:07:36 <woglinde> szsoftware right
May 29 22:07:52 <woglinde> the oe/gpe booth is open since wendsday
May 29 22:07:57 <szsoftware> I'm there, cool.
May 29 22:08:15 <Crofton> the "fatal: Not a valid object name 1" message means you are trying to pull git rev "1"
May 29 22:08:38 <woglinde> szsoftware hehe
May 29 22:08:42 <szsoftware> what is the location of linuxtag?
May 29 22:08:59 <woglinde> messegelände
May 29 22:09:18 <woglinde> http://www.linuxtag.org/2008/
May 29 22:09:55 <szsoftware> Crofton|work: I just learned that because a try of "htc-msm" failed and the look in the log gave me the answer. 
May 29 22:10:23 <szsoftware> 10min from my location :-)
May 29 22:10:28 <szsoftware> ups
May 29 22:13:28 <szsoftware> Crofton|work: Indeed, It seems it works now. It's amazing, I learned so lot!
May 29 22:13:44 <Crofton|work> great
May 29 22:13:46 <szsoftware> :-) NOTE: package linux-kaiser-2.6.24+git9902b6d3437330ff848039d40fafd707da1fa226-r0: task do_fetch: completed
May 29 22:18:45 <ant_> gn all
May 29 23:51:29 <Fidget> Are kernel-module-<mymod> pseudo targets?  If so, is there a way to enable it so that it checks the kernel .config for support for <mymod> as a module/builtin?
May 30 00:15:07 <cdbot2> * * OE Bug 4336 has been created by c.romain(AT)laposte.net
May 30 00:15:09 <cdbot2> * * uucp-1.07-r2: do_install fails when no uucp user on host
May 30 00:15:11 <cdbot2> * * http://bugs.openembedded.net/show_bug.cgi?id=4336
May 30 01:05:04 <Fidget> If I require another image, is there some way that I can exclude a dependency from its IMAGE_INSTALL?
**** ENDING LOGGING AT Fri May 30 02:59:57 2008