IRC log for #maemo-ssu on 20121228

00:02.24DocScrutinizer05IMAP my ass, probably any CIFS mount was smarter than this abomination
00:05.31ShadowJKcifs and nfs mount to maildir emulates imap pretty well :-)
00:06.33*** join/#maemo-ssu kolp (~quassel@212.255.105.160)
00:09.10DocScrutinizer05I wonder if Poettering... NAH *shuddder*
00:11.43Sc0rpiusthere are LOTS
00:11.44DocScrutinizer05admittedly POP3 is kinda old and from a better world, but with a few augments it still works today. IMAP anyway obviously is not the alternative, rather an even worse complementary protocol
00:11.44Sc0rpiusLOTS
00:11.59Sc0rpiusof companies using Google BUsiness and all its private documents in Google Drives
00:12.15Sc0rpiusbut LOTS of companies, not just one, not just 100,  it's THOUSANDS
00:12.37DocScrutinizer05and there a BILLIONS of flies eating shit
00:12.58Sc0rpiusactually the company that I'm working for wants to migrate everything to Google Apps for Business
00:13.09Sc0rpiusbut I don't think they will, not because it's not a good idea (it is)
00:13.09DocScrutinizer05congrats
00:13.13Sc0rpiusit's because it is very expensive
00:14.10Sc0rpiusmost universities around here have moved the students and teachers accounts to Google Apps since it's free for educational entities
00:14.12DocScrutinizer05yeah, probably almost as expensive as a security manager had been, so your company didn't think they need one as well
00:15.06DocScrutinizer05well, selling milk I probably also wouldn't bother about security
00:15.30Sc0rpiusso if you're a student of, I dunno, Colorado State University, your student account is in Google and not in your university racks
00:15.52Sc0rpiusI think almost every university in the states had migrated to Google Apps
00:15.53DocScrutinizer05you think I give a shit about silly students?
00:15.58Sc0rpiusso like I said, everything is going cloud
00:16.24Sc0rpiusit's just an example, I'm not gonna name thousand of companies that have migrated to Google Apps and not having any email/docs/calendar/etc infrastructure at all
00:16.25DocScrutinizer05every idiot is going cloud, agreed
00:17.01Sc0rpiusidiot is having to pay operations manager / engineers $100,000 a year when you can pay Amazon E3 for like $25 a month
00:17.13DocScrutinizer05the companies I worked for last 5 years wouldn't allow you to log in to intranet from at home
00:17.22DocScrutinizer05for security reasons
00:17.46Sc0rpiusthat's stupid, because telecommuting is very important these days
00:18.09Sc0rpiusthat's why most decent companies give very secure VPN solutions (even with USB dongles and whatnot) to their employees so they can log in from home
00:18.17Sc0rpiusmy gf works for HP
00:18.24DocScrutinizer05yeah sure, you're so smart and big business soooo stupid
00:18.25Sc0rpiusand she has a USB dongle + VPN software in her laptop
00:18.38Sc0rpiusok, Hewlett-Packard is not big
00:18.41Sc0rpiusright.
00:18.50DocScrutinizer05no, they're using a VPN
00:19.02Sc0rpiusI'm assuming VPN when you say "log in to intranet"
00:19.13DocScrutinizer05anbd probably some integrity guard software on the laptop
00:19.26Sc0rpiussince you said "INTRANET" by definition that's not Internet-routable, uses private IPs, and you need VPN.
00:19.32DocScrutinizer05actually I know they do, since we use HP stuff
00:19.40*** join/#maemo-ssu kolp (~quassel@212.255.105.160)
00:19.50Sc0rpiuswhere I work for is not that big but we have VPN as well
00:20.01Sc0rpiusa simple PPTP connection
00:20.29Sc0rpiusactually you can't even read mails from home without firing up the VPNs first
00:20.36DocScrutinizer05Sc0rpius: you're boring me. I'm not a silly co-student of yours
00:21.24Sc0rpiusok, but you have to admit you are VERY old fashioned and you are hating what's coming in the future
00:21.29Sc0rpiusand also, you don't want to be educated
00:21.42Sc0rpiusbut you should embrace it
00:21.44Sc0rpiusit's not bad
00:21.54DocScrutinizer05you're very self confident and telling me you can tell the future
00:22.26Sc0rpiusyou can read all about it everywhere you want: slashdot, digg, engadget
00:22.27DocScrutinizer05and I don't give a shit about your notion of what's the future like
00:22.44DocScrutinizer05and if I should or shouldn't embrace it
00:23.24Sc0rpiusok, then go use UUCP to getch your mails.
00:23.28DocScrutinizer05see, I'm still around and living, and using modern technology, probably longer than you are alive
00:23.29Sc0rpiusfetch
00:23.39Sc0rpiusno, sadly you're not older than me
00:23.53Sc0rpiuscome on I've used PDP-11
00:23.57Sc0rpiusyou were a kid by then
00:24.00DocScrutinizer05so what?
00:24.01Sc0rpius:)
00:24.15DocScrutinizer05hah
00:24.42DocScrutinizer05the time the first PDP-8 hit the market I actually wasn't grown up
00:24.50Sc0rpiusstop being so grumpy, embrace IMAP, since it's gonna disappear in the next 5 years and we all are gonna stuck with those stupid web based mail accounts
00:25.15DocScrutinizer05stop giving me advice, I won't take it. I'll beat you with it eventually
00:25.34merlin1991no logon to intranet is rather silly htough
00:25.46merlin1991ah shit I'm like 2 pages late
00:25.50Sc0rpius:)
00:26.00DocScrutinizer05merlin1991: I said "from at home2 and implied "from my own PC"
00:26.55DocScrutinizer05and you won't find a sane company with IP assets that allows loging in to intranet with your own stuff, no matter if from at home or even bring to office
00:27.14DocScrutinizer05au contraire, you might get fired for *bringing* a laptop
00:27.37Sc0rpiuswell my brother works at EMC
00:27.44Sc0rpiusand that's nasty, they hijack all your SSL connections
00:27.53DocScrutinizer05normal
00:28.13Sc0rpiusso you think you're establishing an HTTPS connection to some place but no, you have a middleman finishing your SSL and starting a new one to the final place
00:28.27DocScrutinizer05^^^
00:28.33Sc0rpiusand that stupid "middleman" computer decides if that place you wanna visit is safe or not
00:28.40DocScrutinizer05normal
00:28.49Sc0rpiusit could be your bank account and they are logging EVERYTHING
00:28.57Sc0rpiusat least they tell you when you sign the offer letter.
00:29.29DocScrutinizer05no, they tell you every day when logging in to internet
00:29.39DocScrutinizer05not intranet, INTERnet
00:30.02Sc0rpiusyeah I guess when you open the browser they hijack whatever DNS domain you try with a webpage saying all the rules and an "OK I accept" button at the end
00:30.03DocScrutinizer05i.e when you tell that man in the middle that you want to access internet
00:30.05merlin1991well it is simply silly to use the company network for anything private (what do we have our n900s for?)
00:30.23Sc0rpiusyes exactly
00:30.26Sc0rpiusthis is what he did
00:30.32Sc0rpiushe bought a 4G hotspot
00:30.41Sc0rpiushe bought another SIM card from AT&T
00:31.00Sc0rpiushe puts it there
00:31.12Sc0rpiusand he has a "personal private wifi network" in his cubicle
00:31.44merlin1991ew cubicle, I could never work in such a thing
00:32.06Sc0rpiuswell: office
00:32.16Sc0rpiusI don't know I have never gone to his office
00:33.56merlin1991http://talk.maemo.org/showthread.php?p=1308348#post1308348 anybody got a guess what he/she means by "This is not latest release bug." ?
00:35.34merlin1991btw DocScrutinizer05 did I understand correctly, that you ideal mail setup is a server that has all mails slammed into one place and each client downloads to a local cache applying his/her own filtering/sorting independend of the storage on the server?
00:35.55Sc0rpiuswell looks like the N900 was an extremely popular phone in India.
00:36.16Sc0rpiusI guess hxka means that bug was not present in the previous release
00:36.21DocScrutinizer05merlin1991: exactly
00:38.54DocScrutinizer05since a) my clients are not uniform, b) I want local mails for security and redundancy, easy management of local filters and searching in big volumes via mail client, and for access to all that info while in airplane or otherwise not online
00:39.52merlin1991hm you could (ab)use imap for that, ie in thunderbird you could set up copy filters that filter the mail out of the (imap)inbox into local (non imap) folders
00:40.09Sc0rpiusbut that client has to read that mail first
00:40.12DocScrutinizer05and pop3 serves perfectly for that purpose, while that one openmoko mail server forces friggin imap on me
00:40.15Sc0rpiusand that would be  a problem
00:40.57DocScrutinizer05merlin1991: that's exactly what happens when I download a mail to second PC now, and it vanishes on first PC
00:40.57merlin1991s/filter/copy filter/
00:40.59Sc0rpiusI have all my emails available on a plane using IMAP, that's not the problem.
00:41.16merlin1991ah I had copy in already :D
00:41.35merlin1991DocScrutinizer05: how can it vanish if you copy and not "move" the mails?
00:42.15DocScrutinizer05ooh, copy, yeah maybe that would work, though doubles local storage needed for single purpose to make imap happy
00:42.41merlin1991DocScrutinizer05: I'm sure you can tweak thunderbird to only keep headers of the imap mails
00:42.58merlin1991and not keep a full "local" copy for things in the imap folder
00:43.10DocScrutinizer05I'm not using thunderbird
00:43.24merlin1991well maybe your mail client can too :D
00:43.32merlin1991I guess Kmail then?
00:43.33DocScrutinizer05might be possible for other clients too
00:44.04DocScrutinizer05alas all experiments with that account endanger the mails on server
00:44.50DocScrutinizer05one wrong setting and pooof 250,000 mails gone
00:45.55merlin1991I'd test it with a temp account ;)
00:46.20DocScrutinizer05yeah, temp account on a mailserver with imap :-S
00:46.36merlin1991why not xD
00:46.54DocScrutinizer05because it's friggin annoyance and time consuming
00:47.16DocScrutinizer05that's not a reason, but a concern
00:47.19merlin1991you could say the same about email in general
00:47.38DocScrutinizer05or in short: a major PITA
00:48.28DocScrutinizer05all I planned to do was moving my mail form old to new laptop, while keeping the old one active as backup
00:49.15DocScrutinizer05you could do such migration as your fulltime job
00:49.37DocScrutinizer05when finished, start all over again with next OS release ;-P
00:51.27DocScrutinizer05old PC has kmail1.12.2, on new one I had to downgrade from kmail2 to 1.9.10 since that new shit doesn't work
00:51.51DocScrutinizer05google for akonadi and weep
00:56.04DocScrutinizer05I'll go for copy and removal of mails older than some days
00:56.51DocScrutinizer05though I'd like to nuke that IMAP crap out of our solar system
01:13.58*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
02:07.46*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.142)
03:55.56*** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn)
04:04.27*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
04:59.10*** join/#maemo-ssu Jaded (~jade@Jade.broker.freenet6.net)
04:59.10*** join/#maemo-ssu Jaded (~jade@unaffiliated/jade)
06:47.59*** join/#maemo-ssu Jaded (~jade@unaffiliated/jade)
07:01.59*** join/#maemo-ssu Jaded (~jade@Jade.broker.freenet6.net)
07:01.59*** join/#maemo-ssu Jaded (~jade@unaffiliated/jade)
07:37.52*** join/#maemo-ssu luf (~luf@ip-89-102-208-114.net.upcbroadband.cz)
07:45.34*** join/#maemo-ssu dhbiker (~dhbiker@APN-122-23-200-gprs.simobil.net)
08:01.20*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
08:04.49kerioDocScrutinizer05: btw, i can't believe thunderbird doesn't have a setting to "never delete anything that the server tells you to delete"
08:30.00luffreemangordon: it seems quite ok with libpng NEON backporting :)
08:33.37kerioyay
08:47.29freemangordonluf: good. BTW I was unable to reproduce bluetooth bug here. Though OPP transfer rate was some 1KB/s.
08:50.02freemangordonluf: re NEON lipng - did you find some benchmarks?
08:50.12freemangordon*libpng
08:58.35freemangordonluf: https://gitorious.org/community-ssu/libhildonfm/blobs/master/hildon-fm/hildon-file-system-private.c#line229
09:00.16freemangordonthough we miss obex module in osso-gnome-vfs2, so enabling ^^^ does not help much. but at least file manager accepts obex://[xx:xx:xx:xx:xx:xx/ as location
09:01.21jon_yipv6?
09:01.44freemangordonjon_y: hm?
09:02.17jon_y[xx:xx:xx:xx:xx:xx] looked like ipv6 :)
09:02.40freemangordonno, this is BT address
09:02.50jon_yI was experimenting with dns64 and nat64, latter didn't go too well
09:02.57jon_ymac address?
09:03.00freemangordonyep
09:03.08freemangordonsomething like
09:03.31jon_yI'm curious if the n900 kernel support virtual adapters
09:03.39jon_yany ideas?
09:04.00jon_yerr, I mean vlan
09:04.23freemangordondon;t see why not. though I am not sure the kernel is compiled with that option enabled
09:04.46jon_yok, maybe kp has it
09:04.58freemangordoni doubt
09:05.30jon_ywould it be too much if I asked config.gz be compiled in? :)
09:05.54freemangordonask Pali
09:06.04jon_yok, I will
09:06.26jon_yPali: what do you use for developing uboot and building kp?
09:06.40Palijon_y: ubuntu
09:07.06jon_ystock ubuntu?
09:07.13Paliarm-linux-gnueabi-gcc cross compiler
09:07.31jon_yoh, any debian derivative will do?
09:08.19freemangordonPali: http://lists.maemo.org/pipermail/maemo-developers/2009-October.txt
09:08.26freemangordonsearch for open_folder
09:08.53freemangordonI wonder if that will solve USB hang issue
09:09.32Palijon_y, I'm using Kubuntu (but core packages are same)
09:10.42jon_yok, I assume this is the standard soft float toolchain?
09:11.12jon_yjust add -mthumb for thumb?
09:11.23keriooh, i don't know if the kernel compiles nicely with -mthumb
09:11.47freemangordonforget -mthumb for 2.6.28
09:12.01keriofreemangordon: why, what's the issue?
09:12.09jon_yobviously this is for running with kp
09:12.37jon_yfor kernel, I just want framebuffer
09:12.41keriooic
09:12.53keriowell, KP's got framebuffer i think
09:13.24jon_ywon't it show a bootup log instead of the 5 blinking dots?
09:13.33keriooh, you want fbcon
09:13.42jon_yerr, yeah that
09:14.00jon_yI can't really tell them apart since I haven't actually play with it yet
09:14.36jon_yjust look for fbcon and make it built-in right?
09:15.30jon_yif it is just that, I'm going to install debian NOW :)
09:15.44kerioyou need to boot with the appropriate parameters ofc
09:16.26keriohm, maybe not actually
09:16.32jon_yare they not part of the built in command line?
09:17.01jon_yI'm already using uboot, so it is just cloning the boot config right?
09:17.56keriojon_y: apparently you just have to modprobe fbcon early
09:18.04kerioso i assume that compiling it in gives you an even earlier console
09:18.11jon_yah, even simpler
09:21.00kerioPali: how would one load a module early on in the boot process?
09:22.19jon_yI remember there was an init file that controlled the boot video
09:22.50keriothe boot video is run by hildon-welcome
09:23.15keriothe dots are hildon-startup-progress
09:23.18Palikerio, install kernel-power-settings and add module to /etc/modules
09:23.20kerioboth are uninstallable cleanlyish
09:23.34Palikerio, or create new /etc/event.d/<file> and add here modprobe module
09:24.21jon_ykerio: is there a way to disable both without uninstalling?
09:24.42keriojon_y: sure, change the appropriate files in /etc/event.d/
09:25.44jon_ythanks
09:26.03keriothere was a line you had to disable to uninstall hildon-startup-progress cleanly
09:29.27jon_yhmm, disabled the blinky dots but putting fbcon in /etc/modules don't seem to work
09:30.03jon_ydoes /etc/modules or /etc/event.d/hildon-* run first?
09:30.36keriomaybe you also have to specify that you want the console there
09:31.19jon_yneed to add console=tty0 or something in the kernel command line?
09:32.36keriotty1, if we're to believe backupmenu
09:32.55jon_yok, will try later
09:35.48freemangordonluf: http://maemo.org/packages/package_instance/view/fremantle_ssu_pr1-2_armel/osso-gnomevfs-extra/1.9.6-3/
09:36.03freemangordonobex gnomevfs backend :)
10:29.55*** join/#maemo-ssu arcean (~arcean@aacv246.neoplus.adsl.tpnet.pl)
10:30.45*** join/#maemo-ssu NIN101 (~NIN@p5DD28D03.dip0.t-ipconnect.de)
10:43.59*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
10:45.33jon_yPali: how do I append kernel cmdline?
10:45.44jon_yITEM_CMDLINE seem to mean replace the entire cmdline
10:45.45Palijon_y, where?
10:45.56jon_yin the uboot .item file
10:46.08Paliyes, ITEM_CMDLINE will replace in-build kernel cmdline
10:46.36Paliyou cannot append to cmdline which is hardcoded in zImage
10:46.54jon_yok, co copy /proc/cmdline and modify appropriately?
10:47.10Paliyes
10:47.15jon_yok
10:48.05*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
10:51.39jon_yPali: any idea what ttyMTD is?
10:53.44keriojon_y: tell me if you manage to make it work
10:53.54jon_yok, trying now
10:54.34keriojon_y: apparently ttyMTD,log is for mtdoops
10:54.53jon_ysweet
10:55.12Paliyes ttyMTD is for oops logs
10:55.14jon_yworking, blacks out after the blinking dot kicksin
10:55.20kerio"log" is the name of the mtd partition
10:55.33keriojon_y: neat, what did you change?
10:55.36Pali"console=ttyMTD,log" selects the MTD named "log" for mtdoops.
10:55.42Palihttp://lists.infradead.org/pipermail/linux-mtd/2009-February/024631.html
10:55.46jon_yin uboot, add console=tty1
10:55.55kerioand modprobe fbcon?
10:55.56jon_ythere was already a console=tty0
10:56.01keriok
10:56.06jon_yyeah, and that too in /etc/modules
10:57.20jon_yso, can enough kernel boots destroy the log partition?
10:57.50keriojon_y: enough oopsies
10:57.53freemangordonluf: seems obexftp is fully supported in ossofilemanager/ossognomevfs2, we just need to tweak obex gnomevfs module to use BT4 dbus api
10:58.37jon_ykerio: well, kernel is good now, so no need to do that until I need to debug :)
10:59.07kerioi wonder if that logging increases battery usage
10:59.25kerioi mean, the fb console
10:59.51jon_ywell, probably less than writing to nand
11:01.12keriothere's nothing written to nand
11:01.16kerioexcept in case of an oops
11:01.46Palijon_y, what are you trying?
11:01.53jon_ydoes that include crashing when messing with modprobe?
11:02.02kerioi suppose so
11:02.05jon_yPali: I'm good that fbcon is working
11:02.17jon_ynow I'm just messing about
11:03.19freemangordonluf: http://i48.tinypic.com/1zour12.jpg
11:03.47keriofreemangordon: neat, for obex exploring?
11:03.53freemangordonyep
11:04.12freemangordonwe just need to change dbus api for newer bluez
11:04.29freemangordonon an FOSS module
11:04.32freemangordon*in
11:04.57freemangordonthat one used to be supported in maemo4
11:05.01freemangordonafaik
11:07.13*** join/#maemo-ssu ekze (~nyan@bakaekze.ru)
11:09.59jon_yinteresting
11:10.13jon_ywithout console=ttyMTD,log, no messages at all
11:10.18jon_yneed to double check
11:18.49DocScrutinizer05jon_y: kerio: Pali: you're not going to nuke mtdoops, are you?
11:19.14PaliI did not touch mtdoops
11:19.17keriono one is doing shit doc
11:19.23DocScrutinizer05mtdoops *always* been around, and NO it doesn't need any significant amount of energy
11:20.12Palifreemangordon, do you know state of portrait keyboard?
11:28.38jon_ykerio: what is your /proc/cmdline?
11:29.30kerioinit=/sbin/preinit ubi.mtd=rootfs root=ubi0:rootfs rootfstype=ubifs rootflags=bulk_read,no_chk_data_crc rw console=ttyMTD,log console=tty0 snd-soc-rx51.hp_lim=42 snd-soc-tlv320aic3x.hp_dac_lim=6
11:29.57jon_yok, odd, my boot messages no longer show
11:32.31freemangordonPali: state as in?
11:32.50freemangordonwhat I know that something works in portrait, never tried it
11:46.07jon_ykerio: sadly, the boot messages no longer shows
11:46.41jon_ywhich is to say weird as hell
12:13.14*** join/#maemo-ssu DocScrutinizer51 (~lagrange@openmoko/engineers/joerg)
12:22.02*** join/#maemo-ssu Martix (~martix@4.177.broadband3.iol.cz)
12:24.18*** join/#maemo-ssu kolp (~quassel@212.255.21.58)
12:34.36*** join/#maemo-ssu Estel_ (~Estel@Maemo/community/contributor/Estel-)
12:39.57*** join/#maemo-ssu Estel_ (~Estel@213.158.217.201.pat.umts.dynamic.t-mobile.pl)
12:39.57*** join/#maemo-ssu Estel_ (~Estel@Maemo/community/contributor/Estel-)
12:53.29luffreemangordon: cool :)
12:54.56luffreemangordon: can you be more descriptive what you enabled/installed? I'll add it to my TODO
12:58.03luffreemangordon: libpng - no benchmark. I haven't backported it yet. Just I took a look into sources. However it's just some read_filter support so I don't know if it helps a lot.
13:14.32*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
13:17.46DocScrutinizer05\o/ quiet times are ober, never again 2h silence on any maemo related channel
13:17.54DocScrutinizer05over even
13:23.42freemangordonluf: I installed osso-gnomevfs-extra
13:25.19freemangordonluf: changed #if 0 to #if 1 here https://gitorious.org/community-ssu/libhildonfm/blobs/master/hildon-fm/hildon-file-system-private.c#line229
13:27.03freemangordonand modified om-dbus.c(osso-gnomevfs-extra), function om_dbus_get_dev_list to enable fake devices
13:27.44int_uaDoc: isn't it because of winter?
13:29.11DocScrutinizer05[general notice] regarding channel moderation of users starting with es and ending with l_ PRETTY PLEASE immediately contact any of the channel moderators jaffa, chrashanddie/teotwaki, chem|st, merlin1991, mohammadag. Preferably those that are already online and most recently active. PLEASE do NOT contact me!
13:31.11freemangordonDocScrutinizer05: estel still banned?
13:31.25DocScrutinizer05nope
13:31.32DocScrutinizer05afaik
13:31.32freemangordonaah, ok
13:39.54*** join/#maemo-ssu toxaris (~toxaris@s83-180-246-172.cust.tele2.se)
13:42.15freemangordonluf: maybe it does not worth the effort to backport NEON if it will gain 1%-2% in speed. I think it is better if you find some benchmarks before doing so, so we can assess if it makes sense. I'll try to find too
13:48.52freemangordonluf: https://codereview.chromium.org/5141003
13:49.25freemangordonthis is for 1.2. unfortunately this is neither upstreamed nor verified/tested aiui
13:50.37freemangordonit is said there we should gain 20-25%
13:55.57freemangordonluf: you can check that one too: https://bugzilla.mozilla.org/show_bug.cgi?id=462796
13:55.58povbotBug 462796: was not found.
14:16.51DocScrutinizer05freemangordon: ack
14:16.57DocScrutinizer05:-D
14:22.15*** join/#maemo-ssu arcean_ (~arcean@aacy22.neoplus.adsl.tpnet.pl)
14:23.29*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
14:29.53*** join/#maemo-ssu mark_g44 (~root@095160235235.zabrze.vectranet.pl)
14:31.53mark_g44Hi. Scratchbox.org is offline, also mirror: http://thecust.net/nelisquare/hathor-sdk/ is offline. Is there any working mirror ? I need scratchbox for amd64 (I have deb files on my hdd, but only for i386 :/)
14:37.32luffreemangordon: this neon patch is totally different from upstream :(
14:37.44*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
14:38.07lufI'll try both and we'll see.
14:45.32DocScrutinizer51mark_g44: ping merlin1991 or thedead1440
14:47.19DocScrutinizer05or teotwaki, resp check wedrop.it mirror if SB is there
14:47.45*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
14:48.08DocScrutinizer05mark_g44: we got some backups, but none of them is "public" right now
14:48.12DocScrutinizer05afaik
14:49.09DocScrutinizer05official maemo migration will take a few days longer, and it's unclear yet if SB guys, Nemein, HildonFoundation Board negotiated something for SB future
14:50.24DocScrutinizer05merlin1991: we got SB backuped on your server??
14:50.34DocScrutinizer05oops, one ? is enough
14:51.13merlin1991yea
14:52.21DocScrutinizer05could you share a link for mark_g44
14:52.34merlin1991not publicly avaiable
14:52.45DocScrutinizer05maybe allow ftp?
14:52.53merlin1991no ftp on the server :D
14:53.03DocScrutinizer05or send it as mail attachment? ;-P
14:53.11mark_g44hmm, I'll install 32bit debian on my pc
14:53.34merlin1991hmm how do I create a user for sftp only?
14:53.46int_uamerlin1991: thumbs up for no ftp :)
14:53.49DocScrutinizer05errr, login shell sftp?
14:53.54merlin1991without login
14:54.03merlin1991like drop.maemo.org
14:54.26DocScrutinizer05sorry, no admin here
14:54.35merlin1991ah shit I just realized I only have the scratchbox mirror locally, sorry
14:55.21int_uaBTW, merlin, do you have the backup of this backup?
14:55.24mark_g44ok, I'll install 32bit debian, because I have i386 debs on my hdd
14:56.30DocScrutinizer05sorry guys, need to get some shopping done before council meeting
14:56.33DocScrutinizer05bbl
14:56.45merlin1991hm bit I do have the files in var/cache/apt/archives/ on the server
14:57.14merlin1991mark_g44: which packges do you need? (the sb on the server is harmattan, so there might be some fremantle stuff missin)
14:57.42merlin1991just relalized, that that mean the gcc tolchain is missing then
14:58.00mark_g44packages needed for linaro toolchain gcc 4.7.2
14:58.25mark_g44scratchbox-core scratchbox-libs and scratcbox-devkits*
14:58.59mark_g44around 250MB
14:59.20DocScrutinizer05:-/ no SB on http://maemo-archive.wedrop.it/
15:00.21merlin1991mark_g44: http://cdnm.at/~christian/sb/
15:00.33DocScrutinizer05anyway http://maemo.cloud-7.de/migration-project/backed-up-sites.txt shows you might contact thedead1440 if in a pinch
15:00.33merlin1991ping me when you got them so I can remove them again
15:01.25DocScrutinizer05merlin1991: is that any useful stuff to host on cloud-7 ?
15:01.55merlin1991I have no idea about the license on them
15:01.59merlin1991so I wouldn't do that
15:02.08DocScrutinizer05I don't give a ....
15:02.11merlin1991most of them are probably gpl and I have no source
15:02.16mark_g44merlin1991: Thank You
15:02.19mark_g44downloading now
15:02.20merlin1991well probably all of them
15:02.31DocScrutinizer05mompl, please don't take it down yet
15:02.59*** join/#maemo-ssu andre__ (~andre@ip-84-42-203-5.net.upcbroadband.cz)
15:02.59*** join/#maemo-ssu andre__ (~andre@wikimedia/aklapper)
15:03.13merlin1991mark_g44: I'm currently booting my laptop, should have the missing devikts for fremantle there
15:06.02mark_g44ok, devkit doctools is missing :/
15:06.24mark_g44git and svn
15:09.55DocScrutinizer05merlin1991: I could oull then into  http://maemo.cloud-7.de/migration-project/sb
15:10.05DocScrutinizer05pull*
15:12.42merlin1991mark_g44: atm I'm uploading the rest
15:13.30merlin1991mark_g44: you don't really need git and svn
15:13.41mark_g44merlin1991: Big Thanks
15:13.45mark_g44ok
15:13.54merlin1991they provide git and svn inside sb, but at least the git package is so outdated, that it foobars newer repos
15:13.59merlin1991or throws random errors
15:14.15merlin1991I usually do svn and git stuff from outside scratchbox
15:15.52mark_g44Should I install scratchbox-toolchain-cs2009 and other toolchains for linaro gcc472 ?
15:16.13merlin1991nope you don't need that if you drop in the linaro toolchain
15:16.13freemangordonmark_g44: what are you trying to do?
15:17.00mark_g44ok
15:17.19mark_g44I want only install the newest linaro gcc4.7.2 toolchain on my debian 64bit
15:17.38merlin1991arf upload speed here SUXX
15:18.14merlin1991oh just realized, devkit doctools is 132Mb that could be the reason aswell :P
15:18.35mark_g44hmm
15:18.45mark_g44I'll install 32bit
15:18.50merlin1991nah I've started it
15:18.53merlin1991I'll finish it
15:18.57mark_g44it's no problem for me
15:19.17*** join/#maemo-ssu BCMM (~BCMM@unaffiliated/bcmm)
15:19.22DocScrutinizer05please toss the filenames in PM so I can pull to cloud7
15:20.56freemangordonmark_g44: in case you find it useful: http://wiki.maemo.org/CSSU-thumb_toolchain_setup_%28gcc4.7.2-linaro%29
15:21.06freemangordonnot the latest 4.7.2, but still
15:21.07merlin1991oh shit, up speed 50kb/s
15:21.41merlin1991mark_g44: depending how fast you need this I can upload it tonight
15:22.30merlin1991I didn't realize my up speed here is that bad
15:22.42mark_g44merlin1991: ok, no problem
15:22.46DocScrutinizer05up is always a pain
15:22.54merlin1991yeah but 50kb?
15:22.59merlin1991that's an insult
15:23.00DocScrutinizer05:-/
15:23.19mark_g44freemangordon: I readed it and I had it
15:23.31mark_g44the newest is 2012_09 :)
15:23.38mark_g44from merlin1991 repo
15:24.17merlin1991mark_g44: freemangordon put the files there ;)
15:25.10mark_g44I know :)
15:32.10DocScrutinizer05anyway, really need to run now
15:32.17DocScrutinizer05cya folks
15:59.13freemangordonmerlin1991: 2012_09 is broken
15:59.55freemangordonmaybe I should remove it
16:01.45*** join/#maemo-ssu ShadowX (~l4m3rx@darknet.escom.bg)
16:20.13*** join/#maemo-ssu nox- (noident@freebsd/developer/nox)
16:34.35*** join/#maemo-ssu dhbiker (~dhbiker@APN-122-172-164-gprs.simobil.net)
17:06.06*** join/#maemo-ssu mark_g44 (~mg@095160235235.zabrze.vectranet.pl)
17:11.50mark_g44merlin1991: Now I'm on debian 32bit and installing scratchbox :)
17:15.50*** join/#maemo-ssu arcean (~Arcean@aacy22.neoplus.adsl.tpnet.pl)
18:23.11lufmerlin1991: scponly ("shell" for allowing only sftp/scp and no login)
18:23.37freemangordonluf: did you see my explanations ^^^
18:24.22lufmerlin1991: it also support chroot but it needs more time ...
18:25.15luffreemangordon: I see nothing after my last reply.
18:26.22freemangordonluf: http://mg.pov.lt/maemo-ssu-irclog/latest.log.html#t2012-12-28T15:23:43
18:26.33kerioluf: it needn't to support chroot, really
18:26.39kerioopensshd can do that for you
18:27.06lufI saw that.
18:27.21freemangordonok
18:27.39lufThanks. I'm adding it to my TODO.
18:28.19freemangordonluf: which one? as obex ftp will be lots of job ;)
18:29.50luffreemangordon: obexftp ...
18:30.05freemangordon:D
18:30.07freemangordonnice
18:30.11lufIt'll be nice to browse remote devices using obexftp.
18:30.21lufSo easily ...
18:30.35freemangordonsure. I hope i'll be able to give a hand there
18:31.07lufFeel free to focus on the kernel and glibc + thumb. It's great and huge job.
18:31.25lufI know I'm slower ... but it's not so important ;)
18:32.08freemangordonBTW i tested yesterday with 2 differents stacks (Tosh/IV?) and 3 drivers (toshiba, broadcom, widcomm). no problem with OPP
18:32.31luffreemangordon: it has to be something specific.
18:33.06freemangordonluf: exactly because you are slower, I hope I will be done with kernel/glibc by the time you need help with obexftp :P
18:33.13lufmerlin1991 prepared test environment for me where it fail so I have to find some time and find the reason.
18:33.40luffreemangordon: It's possible. In that case your help will be welcome :p
18:34.11freemangordonI m interested in having working obexftp too
18:34.22freemangordonnext step could be ssh :D
18:34.43lufwhat? ssh over BT? It's working ...
18:34.59freemangordonno, browse ssh in ossofm
18:35.03freemangordonand smb
18:35.20lufUnderstand.
18:35.28freemangordonyou know smb:/// ssh:///
18:35.51freemangordonseems like the code is there
18:35.57freemangordonbut disabled
18:36.33luf:) Maybe there is a good reason for it. We'll see.
18:37.04freemangordonaiui they didn;t finish it because they we out of time
18:37.09freemangordon*were
18:37.18freemangordonthey == nokia
18:37.25luf:)
18:37.56freemangordonand luf(the bluez man) was not there :P
18:38.10luffreemangordon: I'm not bluez man :p
18:38.29lufI'm just lazy man ...
18:38.40freemangordonhehe
18:39.28ShadowJKAlso it used gnome-vfs and upstream dropped smb
18:39.31luffreemangordon: .arch armv7-a - is this ok for N900? from asm code in libpng ...
18:39.33ShadowJKiirc
18:39.42ShadowJK(it was present in Maemo4)
18:39.51freemangordonShadowJK: yeah
18:40.00freemangordonluf: that's perfect
18:40.24freemangordonluf: do cat /proc/cpu
18:41.17freemangordonShadowJK: any clue what is used in modern distributions?
18:41.28freemangordonfor smb in nauilus
18:41.34freemangordon*nautilus
18:41.44ShadowJKnfi
18:44.36*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
18:45.04luffreemangordon: https://codereview.chromium.org/5141003/patch/1/2 - has support for more bpp but I don't know asm to put it into the upstream code :(
18:45.37freemangordonluf: what?
18:45.45freemangordonluf there are 3 patches
18:46.04lufpng_read_filter_row_neon.S
18:46.48freemangordonluf: this patch is against our version
18:46.56freemangordonit should apply cleanly
18:47.18freemangordonluf: search for +--- a/pngrutil.c
18:47.25luffreemangordon: but I trust more the upstream version.
18:47.56freemangordonluf: but you have to backport it
18:48.08luffreemangordon: upstream version has support only for 3bpp and 4bpp (this patch has also 1bpp, 2bpp, 6bpp and 8bpp.
18:48.13freemangordonaah, i got what you meant
18:48.28luffreemangordon: I'm mostly done with backporting.
18:49.02freemangordonluf: well, we risk either ways
18:49.34freemangordonso I think it is safer to use the patch made agains our version
18:49.44freemangordoneven if it is buggy, we can fix it
18:49.44luffreemangordon: I hope upstream is more tested than this chromium patch.
18:50.08freemangordonluf: maybe, but with different version
18:50.16lufIt's not a problem.
18:50.28lufI red the code ...
18:50.31freemangordonalso, neon code is not included in tar.gz on libpng site
18:51.12freemangordonBTW check the last change of that code, it is a couple of months ago
18:51.16luffreemangordon: are you sure about it? I didn't check if it's included.
18:51.23freemangordonyes, I am
18:51.41freemangordonluf: but you'd better check it too
18:52.37freemangordonluf: I a pretty ok with ARM asm, if there is some bug, I think I can find and fix it. there are some 200-300 instructions after all
18:53.34luffreemangordon: ok.
18:53.53freemangordonluf: did you look here https://bugzilla.mozilla.org/show_bug.cgi?id=462796
18:53.54povbotBug 462796: was not found.
18:55.59lufNo valid patch files were found in the attachment.
18:56.16freemangordonluf: seems mozilla is using (plans to use) the same CodeAurora code
18:56.34freemangordonluf: what?
18:56.35freemangordonhttps://bug462796.bugzilla.mozilla.org/attachment.cgi?id=690720
18:57.17freemangordonof course the patch is fo mozilla, but it should be easy to get the real code from it
18:57.20freemangordon*for
18:57.22lufIt can't show diff. I see the code now.
18:57.39lufBTW didn't you mention some problem with vectorized ...?
18:57.59freemangordonluf: only when gcc does it. this is hand-crafted code
18:59.26freemangordonluf: aah, i see, those are not .patch files, but source code
18:59.36freemangordonit is even easier for us then
18:59.39lufBTW the bugzilla code is the same as the previous link (just contains also zlib neon patch)
19:00.11freemangordonexactly
19:00.16*** join/#maemo-ssu arcean_ (~arcean@aacy22.neoplus.adsl.tpnet.pl)
19:00.54freemangordonBTW neon in zlib should work for us too. though I am not sure if libpng is using its own or system zlib
19:02.00freemangordonluf: seems that  CodeAurora code works after all. And as it supports wider range of bpp, i'd say - go fo it :D
19:02.21freemangordonplus it wont need backporting
19:02.58luffreemangordon: ok. I wasted another hour or two of my life :D
19:03.19freemangordonluf: aah, sorry for that :(
19:03.42lufMy wrong decisin ;) So don't worry ;)
19:03.45luf... be happy
19:03.46freemangordonluf: BTW ^^^ are only my thought, feel free to do whatever you want :)
19:03.58freemangordon*thoughts
19:04.21luffreemangordon: I know but it seems it's the better way from my point of view too.
19:04.58freemangordonluf: well if you are convinced too...
19:05.11freemangordoneveryone is happy :D
19:13.33freemangordonarcean: did you see the bugreports on TMO?
19:13.41freemangordonfor h-d that is
19:20.15luffreemangordon: do you know if this "#if defined(__ARM_NEON__)" is defined or I have to define it manually?
19:21.30freemangordonluf: I think you have to define it
19:37.36freemangordonluf: BTW do you mind to create neon branch on gitorious and to push your changes there
19:37.47freemangordoni'll have some free time tomorrow
19:38.00lufWho knows :D
19:38.11freemangordonwho knows what?!?!
19:38.37lufIt's answer for your question ;)
19:39.02freemangordonluf: noone knows if you mind? :D
19:39.10lufsure ;)
19:39.17freemangordonaah, i see
19:39.46freemangordonhowever, if you push your changes tomorrow i'll try to run some benchmarks and tests
19:40.17freemangordons/changes/changes,/
19:40.22lufOk. I have some weird troubles now with compiling so who knows if I push it ...
19:40.32freemangordonluf: does not matter
19:41.06freemangordononce you got sick of it, push the code on gitorious, i'll take it from here
19:41.23lufAssembler source seen but `CCAS' is undefined
19:41.32lufAny hint?
19:41.39freemangordonMFC what is that
19:41.43freemangordonaah
19:41.54freemangordonmaybe you need to include AC_PROG_AS
19:41.57freemangordonor something
19:42.13freemangordonin Makefile.am or who knows where
19:42.25freemangordonin some autotools script
19:42.28lufaaah I see.
19:42.32lufconfigure.ac
19:42.42freemangordonfor example
19:42.46freemangordon:D
19:44.46freemangordonluf: it is AM_PROG_AS
19:45.04lufYes I went throught.
19:45.12freemangordonaah, ok
19:45.25lufThe message was more descriptive.
19:45.36lufBut I can't believe it's so simple :)
19:45.49freemangordonluf: but please. commit .S and changes to .c code in a different commits
19:46.02lufWhy?
19:46.10luf(just curious)
19:46.35freemangordonit is easily readible when you splict code changes in parts
19:46.38freemangordon*split
19:46.59lufas you wish ;)
19:47.05freemangordon*more easily
19:47.08freemangordon:D:D:D
19:47.27lufIf I remember it :)
19:47.37freemangordonwell, it is not critical
19:47.52freemangordondoes it compile now?
19:48.36lufNot fully because some build obstructions. I also add the configure.ac change into diff ...
19:48.53freemangordonok
19:49.05freemangordonis afk
19:49.34lufWhat is afk?
19:50.23lufYes it's go through configure part but I made a mistake in patch ;)
19:51.44freemangordon~afk
19:51.44infobot[afk] Away From Keyboard
19:52.30lufHey it's magic. You're away from keyboard and drop a message here. What input device are you using?
20:03.52*** join/#maemo-ssu infobot (~infobot@rikers.org)
20:03.52*** topic/#maemo-ssu is Maemo Community Seamless Software Update "CSSU" channel, http://wiki.maemo.org/Community_SSU | Known bugs: http://j.mp/communityssu-bugs | Channel logs: http://mg.pov.lt/maemo-ssu-irclog/ | Sources: http://gitorious.org/community-ssu/ | Latest version: Testing(2012-12-25): 21.2011.38-1Tmaemo7.1; Stable(2012-12-24): 21.2011.38-1Smaemo5 | Meeting: 27th 1900UTC
20:03.53*** mode/#maemo-ssu [+v infobot] by ChanServ
20:13.11DocScrutinizer05yohoho folks, how's life?
20:21.36keriofairly decent
20:22.27lufIt's great next year nearly behind me :D
20:27.25*** join/#maemo-ssu dhbiker (~dhbiker@APN-122-153-41-gprs.simobil.net)
20:38.32lufHmmm any idea how to benchmark libpng with enabled NEON optimization?
20:40.08keriofreemangordon: is zlib also thumbified?
20:40.18lufIt seems it's working here :)
20:40.43lufBut no measurement ...
21:00.22luffreemangordon: I push the changes to the git (to the master).
21:00.38luffreemangordon: ... libpng ...
21:01.38luffreemangordon: I checked the presence of the _neon function in the .so files
21:01.47keriofreemangordon: libpng uses zlib, you should thumbcompile that too
21:02.19lufkerio: don't be so hurry ;) I'm going to check the NEON opt in zlib ...
21:02.30keriowtf, zlib has NEON too?
21:02.37kerioisn't it for floats
21:02.41lufhttps://bugzilla.mozilla.org/show_bug.cgi?id=462796
21:02.42povbotBug 462796: was not found.
21:02.58kerioluf: <3
21:03.37lufkerio: what is less then 3? :D
21:04.02kerioyour ability to accept gratitude without snarky remarks
21:15.14*** join/#maemo-ssu NIN101 (~NIN@p5DD28D03.dip0.t-ipconnect.de)
21:33.31DocScrutinizer05idly wonders if mgedmin could teach povbot to parse URLs and not nag with nonsense when a URL evidently isn't pointing to bugs.maemo.org
21:39.20freemangordonluf: why in master?
21:39.45freemangordonkerio: neon is SMD
21:39.50freemangordonno matter int/float
21:39.54freemangordonsame as MMX/SSE
21:39.59kerioi see
21:47.33luffreemangordon: because it's cool.
21:47.55luffreemangordon: https://build.pub.meego.com/package/view_file?file=zlib-1.2.7-adler32_vec_kaffeemonster.patch&package=zlib&project=home%3Astskeeps%3Acheck-zlib&rev=c0999fc2e13c75f801849437bed4f9ea
21:48.46luffreemangordon: zlib upgrade + neon?
22:09.33*** join/#maemo-ssu ruskie (ruskie@sourcemage/mage/ruskie)
22:24.35*** join/#maemo-ssu jade (~jade@Jade.broker.freenet6.net)
22:24.35*** join/#maemo-ssu jade (~jade@unaffiliated/jade)
23:13.02*** join/#maemo-ssu ekze (~nyan@bakaekze.ru)

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