IRC log for #openezx on 20090429

00:06.44*** part/#openezx p3t3r__ (n=peter@wlan.net-you.de)
01:24.24*** join/#openezx SuN (i=unices@82-170-225-106.ip.telfort.nl)
01:42.10*** join/#openezx flyhorse|linux (n=flyhorse@116.18.254.216)
03:10.32*** join/#openezx arne_ (i=nobody@dslc-082-082-064-184.pools.arcor-ip.net)
06:03.48*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
06:20.32*** join/#openezx darion76 (n=darion@plant.ogneupor.com)
06:28.24*** join/#openezx WyrM (n=wyrm@201.58.130.45)
06:38.51*** join/#openezx l403 (n=l@85.132.159.239)
07:13.04*** join/#openezx trod (n=trod@59.163.114.3)
07:13.34Hopscotchgood morning
09:07.21*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
09:14.41yanghey ao2 come va ?
09:14.56ao2hi
09:15.15ao2I am fine, yang, many thanks. How are you doing?
09:15.52yangI am fine aswell
09:16.09yangHas Italia recovered from the Earthquakes ?
09:17.16yangmabye some of you would like to expand the text in https://secure.wikimedia.org/wikipedia/en/wiki/OpenEZX ?
09:21.55yangpings sabrod
09:22.00ao2yang, about earthquake, weather is still causing some little problems, but people should be safe now. The really big problem is the reconstruction, the long hand of dark economies (read also Mafias :) ) could take big advantages from the disaster...
09:22.53yangao2: In Slovenia they still haven't recovered all building from the earthquake in 1999
09:23.37yangao2: the common practice is, that you need your apartment insured for eartquake (comes about 50 EUR on top of the apartment insurance per year)
09:23.43yangin Slovenia
09:25.03ao2I don't know how insurances on houses work here
09:26.37ao2yang, I am giving  a look at the wikipedia article, I think it is fine so brief, just make the word OpenEZX an hyperlink if you can
09:27.01sabrodPong
09:27.02yangAs i can remember there was a state funding of only about 5000 EUR per house for the earthquake disaster, and you know this is not enough to restore the collapsed houses
09:27.20sabrodHi Ao2, yang.
09:27.47yangao2: hyperlink, what exactly do you mean ?
09:28.12ao2yang, make it point to www.openezx.org
09:28.28yangit does, under reference no. 1
09:29.02yangao2: you cannot make hyperlinks inside the article, only on the bottom (External links)
09:29.43ao2ah, wikipedia only allows internal links inside the article, I didn't know that
09:29.48yangyes
09:30.22yangmaybe sabrod has some ideas, how to expand the article
09:30.43tmztwhat is the https link for?
09:31.07yangtmzt: eh, just my https connection, I use it to securelly log into wikipedia
09:31.19tmztI see
09:31.23sabrodHeh, it's already good... ( I don't think this page needs a buggy script to be pasted on ;-) )
09:31.43tmzthaven't seen that before
09:32.15yangits the same what I am trying to suggest to Hopscotch to enable secure connection for OpenEZX website
09:33.13yangit just needs a SSL cert
09:34.04sabrodwebsite is hosted somewhere else, but speaking of the wiki, maybe the mediawiki... yes could be updated.
09:34.17yangsabrod: not its an apache-related thing
09:34.52yangsabrod: I ment for wiki logins yes, not for the main site
09:37.17sabrodAh... well it's a good idea. I was speaking of the update for fancy content, that are not allowed yet. Probably we can deal with that. And adding the SSL cert is more important and also easier I suppose. :)
09:38.25yangsabrod: do you have shell access to the wiki server ?
09:39.01*** join/#openezx gDD (n=gDD@125.116.33.58)
09:39.06sabrodNo... I don't...
09:39.36yangsabrod: I think it's Hopscotch who has that, but I never manage to get him to respond
09:40.39sabrodNote that I think I can manage to find how to make the thumbnails to link to a different page than the upload image one, without it, but I just didn't found yet.
09:41.04yangsabrod: yes its possible
09:41.14yangsabrod: consider asking on #mediawiki
09:41.43sabrodYes...
09:42.23sabrodLets do that.
09:58.56*** join/#openezx florian_kc (n=fuchs@port-217-146-132-69.static.qsc.de)
10:04.48tmztWyrM: reading list, should pcap-keys work on motoq or will it need changes?
11:14.35*** join/#openezx jamesmck (n=james@gimel.esc.cam.ac.uk)
11:46.49sabrodk. basicly image linking on a mediawiki needs a plugin or 1.14, so there is a server task to do.
11:47.32sabrod/nick sabrod|away
11:47.58sabrod|away:)
11:48.38*** join/#openezx ade|desk (n=adavey@195.153.131.254)
11:59.22*** join/#openezx alain2210 (n=alain@193.50.159.2)
12:01.20*** join/#openezx gmzhuo (n=gmzhuo@118.114.148.133)
14:23.34WyrMtmzt (pcap-keys) should work.
14:23.39WyrMhi all :)
14:30.28gmzhuohi WyrM
14:30.37mickeylmorning wyrm!
14:30.50WyrMhi gmzhuo, mickeyl. :)
14:30.51gmzhuoisp1583 dma mode works now,but it's buggy
14:31.05mickeylremind me... what is that chip for again?
14:31.17WyrMgmzhuo: what was wrong? And how its buggy?
14:31.32gmzhuousb2.0 on e6,e2
14:32.09gmzhuoWyrM:wrong,the dma src and dst is reverted by me :)
14:32.44mickeyloh, they got usb2?
14:32.46gmzhuobuggy is not stable when use dma,maybe is resource conflicts
14:32.46mickeylamazing
14:33.16gmzhuoyes,usb2 device
14:34.03gmzhuohow about fso for openezx now?
14:36.48mickeyllooks pretty good in general, device and modem working fine for phone calls, gprs is broken (pending kernel support?), suspend/resume is being worked on (kernel)
14:40.03mickeylwhat's the last word on gprs from the kernel experts?
14:40.43gDDFSO image looks so NICE, nice future:) http://www.freesmartphone.org/index.php/Image:Fso-architecture.png
14:42.55mickeyl:)
14:43.04mickeylyes, the future could be bright
14:43.16mickeyljust need to fight for some more truly open hardware
14:43.28mickeylthe software part of the stack will come along nicely
14:45.20*** join/#openezx l403 (n=l@85.132.159.239)
14:45.40ao2mickeyl, as soon as I find some time I could work on gps again, I'd got for gpsd, does it work ok with fso? I read that gpsd has dbus hooks now.
14:46.14ao2s/got/go/
14:46.50mickeylah right
14:46.54mickeylok, so the deal is that:
14:47.08mickeylyou would need to add a GpsDevice class in FSO
14:47.22mickeyltailored for EZX
14:47.58mickeylthat one gathers the data out of the device and feeds it to FSO. Then you use the usual GPS dbus API to access it (org.freedesktop.gypsy)
14:48.17mickeylthat way we can use things like tangogps or navit on ezx
14:48.25mickeylwithout needing special device support other than in FSO
14:48.47ao2ok, adding those lines to my notes
14:50.47mickeylwhen you're there, we can talk about how to integrate with python. we can do it as a small C module which we glue into python somehow, so you don't have to reimplement what you have
15:02.37*** join/#openezx james__ (n=james@gimel.esc.cam.ac.uk)
15:04.58*** join/#openezx jamesmck (n=james@gimel.esc.cam.ac.uk)
16:55.43*** join/#openezx stefan_schmidt (n=stefan@p5B03691E.dip.t-dialin.net)
17:09.33*** join/#openezx stefan_schmidt_ (n=stefan@p5B0362B4.dip.t-dialin.net)
18:22.10*** join/#openezx l403 (n=l@85.132.159.239)
18:26.41*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
18:36.19*** part/#openezx ade|desk (n=adavey@195.153.131.254)
18:57.50*** join/#openezx flo_lap (n=fuchs@f048048252.adsl.alicedsl.de)
19:02.01*** join/#openezx alain2210 (n=alain@AMontsouris-151-1-11-139.w82-123.abo.wanadoo.fr)
19:42.40*** join/#openezx stefan (n=stefan@p4FD47259.dip.t-dialin.net)
19:42.43stefanhi
19:43.03stefangit clo
19:43.04stefanpenezx.git
19:43.11stefanremote: fatal: Out of memory, malloc failed
19:43.17stefanhmmm
19:43.20stefan:(
19:44.09ao2stefan, try cloning linus' tree and then using it as reference whan cloning openezx one
19:45.30ao2if this issue is due to the huge topgit history, to fix that issue I guess we should export topgit patches, create a fresh repository and recreate the topgit structure on it.
19:48.29sabrodI just failed to broke my computer!
19:49.27sabrodstefan: you should git clone --reference git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-2.6-stable.git git://git.openezx.org/openezx.git
19:50.26stefanim using this http://wiki.openezx.org/Kernel_hacking_-_linux-2.6
19:50.31stefanis it too old ?
19:50.46sabrodThis could work and use less memory on your computer. Topgit is a little memory ogre.
19:50.57ao2stefan, it is how it should be :) sorry
19:51.22stefanno problem but im not all used to using git and topgit etc
19:51.37sabrodWell, this work sometimes here with 2Gig ram,  when I just booted ;)
19:52.00ao2yang, do you have any experience with xen?
19:52.01stefanhmm it seems to work now
19:52.09sabrodIn fact I do:
19:52.11sabrodgit clone git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-2.6-stable.git
19:52.13stefani just tried the frst thing again
19:52.16sabrodthen
19:52.17sabrodgit remote add openezx git://git.openezx.org/openezx.git && git pull openezx
19:53.13ao2eh sabrod this way you don't have external reference which you can losse if you delete linus tree, I think is is saner than how I do :)
19:53.34sabrodWell, yes, here it's been a long time I couldn't... btw, I never used the --reference command, I use the second method...
19:53.56sabrodOh, that cool! :)
19:55.01stefanhow does topgit come into play with your setup , sabrod
19:55.12stefani dont understand the reason for topgit anyway
19:55.42ao2stefan, we use topgit to seaprate patches by topic so it easier to send them for upstream inclusion
19:55.54ao2if you just want to _use_ the code you don't need topgit
19:56.02stefanok
19:56.10ao2if you want to edit it and send us the changes, they you could
19:56.21ao2s/they.then/
19:56.24ao2s/they/then/
19:56.46ao2eh, thanks purl, I busted you!
19:56.56stefanao2 the point is i have some time now and want to sttart again and this time hopefully get the camera to work
19:58.24ao2stefan, if topgit make you feel unconfortable then don't use it and send us a big patch when you think you've something it is worth sharing
19:59.11ao2just be sure to checkout ezx/current when you hack, because otherwise you don't get all openezx changes
19:59.58sabrodHeh, it's also because I only _use_ the code that my way of doing the checkout is saner... ;)
20:01.18ao2sabrod, it should be equivalent, I can still call topgit after your steps
20:01.35ao2well, without should, it is equivalent
20:02.46stefanerror: pathspec 'ezx/current' did not match any file(s) known to git.
20:03.02sabrodOh, yes, I know, but I will not risk me at this...
20:03.52ao2stefan, is it listed with "git branch" ?
20:03.52stefanim on branch master now
20:04.03stefanonly master
20:04.15ao2then you still don't have openezx objects
20:04.40stefanok i used tg remote --populate thing now
20:04.57sabrodbtw, ao2, did you asked the topgit developers, if it wasn't a problem on topgit?
20:05.07ao2sabrod, not yet
20:05.13stefanok now i am on ezx current
20:05.21sabrodIt seems a lot, I remember, wyrm had 7Gig ram used, or so, last month... :)
20:05.25stefando i need to update anything again ?
20:05.33ao2stefan, then I forgot one step, sorry :)
20:05.38stefannp
20:05.48WyrMstefan: 3.8gb :)
20:06.17stefani want to use a binary toolchain which one should i try
20:06.21ao2stefan, no need to update after populate
20:06.57ao2stefan, try the one suggested on the wiki first
20:07.15stefanThe recommended toolchain for building the openezx kernel and related binaries (see Kernel_hacking_-_linux-2.6) is the OpenEmbedded toolchain.
20:07.20stefan?
20:07.44stefanok thanks  hope to get it compiled
20:07.53stefanlast time it was arne who compiled it
20:08.08stefanbut we never updated the sources
20:08.10ao2I see, but there should be a link to a prebuilt one AFAIR
20:08.16stefanand now our dev computer is a mess ;)
20:08.31stefanyes i have the prebuild one already available
20:08.34stefanill try it
20:08.55ao2http://www.angstrom-distribution.org/unstable/toolchain/
20:09.02ao2ok
20:09.14stefanhere it says i should never talk to wyrm about openembedded ;)
20:09.25stefanalso it uses a different .config
20:09.41stefanmake ARCH=arm CROSS_COMPILE=/usr/local/angstrom/arm/bin/arm-angstrom-linux-gnueabi- zImage
20:09.50stefancan i do this ?
20:09.55WyrMstefan: info is outdated. :)
20:10.11stefansorry i was getting confused ;)
20:10.40WyrMOE is no longer patching our kernel and making me crazy with user support ;)
20:10.55stefanok
20:12.09ao2but still, if you build it yourself you have a little more control
20:17.02stefanyes i know
20:17.59yangao2: I have no xen experience, only linux-vserver
20:17.59stefanwe did that before , however arne did it on a self built gentoo system and it really took a long time
20:31.12alain2210Hello
20:32.22alain2210WyrM: I have read the history of IRC of evening of april 27.
20:32.43alain2210you says that you have found the bug about bp.
20:32.53WyrMalain2210: yes.
20:33.01WyrMand I sent instructions to your email.
20:33.02WyrM:)
20:33.29alain2210sorry, I do not read this mailbox every day.
20:33.35alain2210I will read it.
20:34.03alain2210But I have tried you patch seen on irc. and I still have problems.
20:34.22alain2210the patch makes a lot of changes.
20:34.48alain2210but what corrects the bug ?
20:35.54stefando i need to make modules separately ?
20:37.18ao2stefan, not strictly needed
20:41.32stefancool it boots
20:41.47stefanhow do i get the modules on the system again
20:52.58WyrMalain2210: the mail I sent is with instructions on how to fix the bug only.
20:53.07WyrMthe patch on irc is incomplete.
20:53.12alain2210I'm trying.
20:53.15WyrMI will finish clean up later.
20:54.04alain2210I does not understand exactly the problem. I will try.
20:56.37alain2210WyrM: Thanks a lot. It works !
21:05.24*** join/#openezx l403 (n=l@85.132.159.239)
21:19.25*** join/#openezx l403 (n=l@85.132.159.239)
21:36.51*** join/#openezx l403 (n=l@85.132.159.239)
21:49.03alain2210WyrM: very good work. even if I still does not understand. I'm very happy.
21:50.30WyrMalain2210: good.
21:50.36WyrMnow it supends resumes ok?
21:51.01WyrMalain2210: you can also remove all the UHCR foo...
21:51.12WyrMits a layering violation.
21:51.30WyrMand the ohci-pxa27x driver does similar stuff. ;)
21:52.26WyrMin fact, I just started the clean-up on the mux_usb.c driver.
21:52.34WyrMand I removed more than 300 lines so far. ;)
21:53.07WyrMconsidering the driver was 800 lines... Its a lot.
21:55.56alain2210Yes it works.
21:56.38alain2210But what was the interest of bvd_ipc_disconnect ?
21:57.20alain2210It was not done only to broke the driver ;-)
21:58.50alain2210300 lines over 800 ! is the driver needed  ? ;-)
22:00.08WyrMalain2210: it looks like the driver was touched by a lot of people inside motorola.
22:00.18WyrMevery one reinvented the whell a bit ;)
22:00.51alain2210WyrM: perhaps the kernel publish is not the one used.
22:01.18WyrMalain2210: it is.
22:01.56alain2210how can you know that ?
22:02.28WyrMalain2210: I can't be sure.
22:02.49alain2210and how do you have found the solution ?
22:02.54WyrMbut I hope motorola did not deliberately released different source.
22:03.06alain2210I hope also.
22:03.16WyrMalso, the first source release contains "motorola confidential" headers..
22:03.22WyrMand it is complete, for sure :)
22:03.37WyrMI looked at bvd_ipc pointers.
22:03.46alain2210and ?
22:03.56WyrMon disconnect, even bvd_ipc being a global, it was not being used.
22:04.03alain2210yes
22:04.25WyrMthen I checked the addr returned by the intf_get_data, and saw that it doesn't match the bvd_ipc global.
22:04.34WyrMand it should.
22:04.35alain2210ok.
22:04.44WyrMthen I just fixed it the easiest way possible...
22:04.49WyrMusing the global :)
22:06.00alain2210so to understand why it works on a780 but not on e680, one would have to search why
22:06.05WyrMthe disconnect was overwriting some stuff, and making it break elsewhere...
22:06.21WyrMthese overwrites always breaks elsewhere.
22:06.22alain2210usb_get_intfdata(intf) does not gives the good address ?
22:06.29WyrMno.
22:07.37alain2210the code is better now.
22:07.51WyrMis it stable for you now?
22:08.28WyrMafter the clean-up it should be much more stable and cleaner..
22:08.34WyrMeasier to read at least ;)
22:08.34alain2210yes.
22:09.04alain2210using cmer, I have standby it a long time.
22:09.16alain2210(yes I will try sleep).
22:09.45WyrMsleep may still be broken..
22:09.51WyrMbut in this case, not on this driver.
22:10.02WyrMbecause of gpio sleep states.
22:10.10WyrM(which are not used on standby mode)
22:11.51alain2210But it will be easier to test now.
22:12.15WyrMwell, I always use sleep to test.
22:12.27WyrMwith or without the core pm_test
22:12.43WyrMbasically it has the same effect... ;)
22:14.10alain2210yes sleep with pm_test to core is also working.
22:20.12stefana2o , still there
22:20.43stefani successfully modprobed lots of camera modules ;) and now i have a video  and video9 device
22:21.02stefanany advice how i could test it  now ?
22:21.09stefanthat is video0 not video 9
22:24.39alain2210WyrM: very stable now. goodbye.
22:34.41ao2stefan, you can use 'motion' from openembedded, or cross compile v4l2-apps, you find them at linuxtv.org somewhere
22:35.51ao2leaving now, see you tomorrow
22:36.28stefanthanx
22:37.15stefanby the way Thank you whoever fixed the touchscreen keyboard repeat speed problem ;)
22:38.02ao2you're welcome. cat #openezx | sed -e 's/whoever/WyrM/g'
23:21.39*** join/#openezx l403 (n=l@85.132.159.239)
23:33.37*** join/#openezx l403 (n=l@85.132.159.239)
23:36.36*** join/#openezx bjk (n=bjk@244.sub-75-201-79.myvzw.com)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.