| 00:36.17 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 01:13.50 | *** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au) |
| 01:49.21 | *** join/#maemo-ssu discopig (~discopig@unaffiliated/discopig) |
| 02:06.36 | *** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn) |
| 02:38.26 | *** join/#maemo-ssu Milhouse (~Milhouse@Maemo/community/contributor/Milhouse) |
| 04:50.36 | *** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl) |
| 05:06.07 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 06:24.03 | *** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali) |
| 06:48.30 | freemangordon | Pali: hi, it looks as I was able to port back camera to clock framework, where is that "new" firmware I need? |
| 06:49.12 | Pali | freemangordon: https://gitorious.org/omap3camera/camera-firmware |
| 06:49.14 | Pali | last git commit |
| 06:49.18 | freemangordon | thanks |
| 06:49.27 | Pali | previous git commit is 2.6.28 version |
| 06:49.49 | Pali | btw it is not firmware, but kernel C structure with some data |
| 06:50.09 | Pali | and maybe it could be good idea to include these data directly to driver :-) |
| 06:50.15 | Pali | (if driver start working) |
| 06:50.19 | freemangordon | yeah |
| 06:54.42 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 07:15.11 | *** join/#maemo-ssu sixwheeledbeast (~paul@host-78-145-81-34.as13285.net) |
| 07:24.59 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 07:41.59 | *** join/#maemo-ssu discopig (~discopig@unaffiliated/discopig) |
| 08:49.00 | *** join/#maemo-ssu Sohil876 (~Sohil876@106.66.15.210) |
| 09:00.30 | *** join/#maemo-ssu Martix (~martix@123.225.broadband2.iol.cz) |
| 09:20.35 | *** join/#maemo-ssu sunny_s (~sunny_s@business-092-079-020-027.static.arcor-ip.net) |
| 09:29.13 | chem|st | freemangordon: I actually wanted to talk with you what we can do about the gap between them |
| 09:30.15 | chem|st | there are several things that, if we keep "freedom of choice", will never make it to cssu-s as noone ever cares about implementing a choice |
| 09:30.46 | chem|st | like camera and stuff |
| 09:31.11 | freemangordon | chem|st: sure, but at least the sruff that doesn't restrict that "freedom of choice" (and considered stable) could be upgraded |
| 09:31.18 | freemangordon | stuff even |
| 09:31.25 | chem|st | people using stable ask about stuff in testing and end up installing testing as I can only tell it wont make it in the near future |
| 09:32.23 | chem|st | yeah I need to put a new stable together in the next weeks with the stuff we have but we need to think about the general setup of cssu |
| 09:32.40 | freemangordon | chem|st: and honestly, maybe I am not the exact person to discuss "freedom of choice" with, as I see the whole CSSU as a relaxed continuation of Nokia's policy |
| 09:33.12 | chem|st | freemangordon: I would just put testing to stable at some point |
| 09:33.26 | freemangordon | :nod: |
| 09:33.28 | chem|st | like now |
| 09:33.38 | freemangordon | chem|st: nnah |
| 09:33.45 | chem|st | hehe |
| 09:33.53 | freemangordon | there are stuff in -T which is not stable enough afaik |
| 09:34.39 | chem|st | I mean the last known good |
| 09:35.13 | freemangordon | chem|st: but srsly, both you and merlin1991 should agree about the policy re nicocam and such. I know this was agreed about 2 years ago (additional repos, etc) but so far nothing done |
| 09:36.02 | chem|st | he is ok with me putting it just into stable... I have to live with the fingers pointed and the people crying |
| 09:36.08 | freemangordon | and I don;t see anyone workong on that |
| 09:36.45 | freemangordon | chem|st: I am fine with that, not to say i think this is the right thing to do |
| 09:36.45 | chem|st | how many are working on cssu? |
| 09:37.15 | chem|st | ... |
| 09:37.27 | freemangordon | me, Pali, luf, arcean. and _ade_ works on adeclock, but this is still not in cssu |
| 09:37.39 | freemangordon | I hope I don;t miss someone :) |
| 09:38.46 | Pali | it is in cssu-devel :-) |
| 09:39.12 | freemangordon | Pali: yeah :) |
| 09:39.37 | freemangordon | chem|st: anyway, something should be done, esp in light of neo900 |
| 09:40.01 | chem|st | was a rethorical question.... |
| 09:40.38 | freemangordon | we didn;t discuss it with Pali, but my impression is that both of us aim CSSU as being the fremantle distribution to be used on neo900 |
| 09:40.44 | freemangordon | chem|st: ooh :) |
| 09:40.52 | freemangordon | anyway, I counted them :D |
| 09:40.57 | chem|st | ~5 |
| 09:41.31 | freemangordon | or less, as arcean is focused more or less on h-d |
| 09:41.57 | freemangordon | Pali: is my impression correct? |
| 09:42.13 | Pali | why not? |
| 09:42.19 | freemangordon | :nod: |
| 09:45.01 | freemangordon | anyway, gtg |
| 10:09.36 | *** join/#maemo-ssu joshgillies (~josh@ppp108-14.static.internode.on.net) |
| 10:17.36 | *** join/#maemo-ssu sunny_s (~sunny_s@business-092-079-020-027.static.arcor-ip.net) |
| 10:37.44 | *** join/#maemo-ssu lizardo (lizardo@nat/indt/x-rddxhwjgxgtclchh) |
| 10:50.01 | *** join/#maemo-ssu dos11 (~dos@unaffiliated/dos1) |
| 11:44.00 | oooaaaooo | hi guys what is the default root password on the n900 |
| 11:44.13 | oooaaaooo | for su access |
| 11:44.26 | DocScrutinizer05 | ((<chem|st> people using stable ask about stuff in testing and end up installing testing as I can only tell it wont make it in the near future)) aren't cssu-s users free to apt-get install anything from cssu-t they want? |
| 11:45.06 | DocScrutinizer05 | if they're not than we have an elementary problem which needs to get solved |
| 11:45.17 | DocScrutinizer05 | s/than/then/ |
| 11:46.58 | Pali | oooaaaooo: rootme |
| 11:47.07 | DocScrutinizer05 | freemangordon: there BEEN NO "additional repos" involved in what merlin1991 and me introduced 2 years ago to solve the "freedom of choice" issue. It's a simple problem of packaging |
| 11:47.36 | DocScrutinizer05 | oooaaaooo: nope, there is no default password |
| 11:48.04 | *** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au) |
| 11:48.19 | Pali | DocScrutinizer05: look at /etc/shadow after flashing PR1.3 |
| 11:48.20 | DocScrutinizer05 | oooaaaooo: you install rootsh and then use command "root" to gain root permissions |
| 11:48.28 | *** join/#maemo-ssu WielkiTost (~dos@unaffiliated/dos1) |
| 11:48.43 | jonwil | DocScrutinizer05: ping |
| 11:49.02 | oooaaaooo | DocScrutinizer05: so i dont need to su dpkg ? |
| 11:49.33 | Pali | oooaaaooo: if you do not have R&D mode enabled, you need to install rootsh package via Application Manager |
| 11:49.42 | oooaaaooo | Pali: thanks |
| 11:49.42 | DocScrutinizer05 | oooaaaooo: for dpkg and apt-get etc you gain root shell via cmd "root" |
| 11:50.13 | Pali | but for it you need to install rootsh first |
| 11:50.16 | DocScrutinizer05 | oooaaaooo: "su" is "broken" on maemo, thanks busybox |
| 11:50.39 | oooaaaooo | DocScrutinizer05: ok good to know |
| 11:50.39 | Pali | su is symlink to busybox, and symlinks do not have setuid bit |
| 11:51.00 | Pali | so su working only if you start it as root :-) |
| 11:51.18 | DocScrutinizer05 | oooaaaooo: so - install rootsh via HAM, then use cmd "root" to gain a root shell |
| 11:51.41 | jonwil | so guys, do you think wiki.maemo.org is the best place for Neo900 software things or is there a better place for it? Specifically, stuff about which binary packages we can reuse and which binary packages we need to do something different with as well as the dbus stuff for the cellular services daemon stuff |
| 11:51.50 | DocScrutinizer05 | jonwil: pong |
| 11:52.37 | DocScrutinizer05 | jonwil: that's not exactly about Neo900, it's about fremantle porting, and that clearly goes to wiki.m.o |
| 11:52.51 | jonwil | Its about " |
| 11:53.04 | jonwil | its about "porting Fremantle to things based on some hardware assumptions about the Neo900" |
| 11:53.15 | DocScrutinizer05 | :nod: |
| 11:53.25 | DocScrutinizer05 | wiki.m.o perfect location for this |
| 11:53.58 | jonwil | I will create http://wiki.maemo.org/Cellular_Services_Daemon for the CSD and CSD dbus stuff |
| 11:54.39 | DocScrutinizer05 | please consider creating a category "porting/" |
| 11:55.25 | jonwil | ok, makes sense |
| 11:55.28 | DocScrutinizer05 | wiki admins have power to clean up by moving pages around, but better start in right location right away |
| 11:56.16 | jonwil | Also do we know about audio hardware on Neo900? Will it use N900 sound chip? N900 speakers? |
| 11:56.31 | jonwil | or will it use the same hardware as GTA04 for that? |
| 11:57.40 | DocScrutinizer05 | documenting existing dbus calls is clearly a general maemo topic, thus your location as suggested by you is fine for that. Considerations about how much we need to reimplement are not of any interest to the general maemo developer and thus should go to a category "porting/CSD" |
| 11:58.07 | *** join/#maemo-ssu sunny_s (~sunny_s@business-092-079-020-027.static.arcor-ip.net) |
| 11:58.30 | DocScrutinizer05 | jonwil: audio is the most annoying subject in Neo900, and not yet finalized. But most definitely will _not_ use identical chipset like N900 |
| 11:58.38 | jonwil | ok |
| 11:58.44 | DocScrutinizer05 | possibly it's completely different to N900 |
| 11:58.58 | DocScrutinizer05 | except speakers themselves |
| 11:59.03 | DocScrutinizer05 | ;-) |
| 11:59.57 | DocScrutinizer05 | we're still about to wonder why the heck nobody else (pandora, N900... etc) used twl4030 audio function blocks |
| 11:59.59 | FatPhil | Peter Ujfalusi once gave a presentation on the functionality of the TI sound core. He explained that every single possible mode of operation it had was either fundamentally flawed by design, or just plain buggy. |
| 12:00.17 | oooaaaooo | is maemo a rolling release distro or do i have to manually update to the most recent LTS? |
| 12:00.23 | DocScrutinizer05 | aaah thanks, that explains this :-D |
| 12:00.44 | DocScrutinizer05 | oooaaaooo: basicaly maemo is NO rolling release |
| 12:01.03 | jonwil | I will create http://wiki.maemo.org/CSD for discussing the architecture of the CSD and the dbus calls |
| 12:01.06 | DocScrutinizer05 | oooaaaooo: you're supposed to use HAM to do release updates |
| 12:01.23 | oooaaaooo | DocScrutinizer05: ok so its all in ham |
| 12:01.28 | DocScrutinizer05 | oooaaaooo: do NOT do apt-get upgrade or even dist-upgrade! |
| 12:01.34 | FatPhil | But HAM is teh suck. :-( |
| 12:01.38 | DocScrutinizer05 | sure |
| 12:02.02 | DocScrutinizer05 | alas it's the only package manager that knows about Nokia's package groups they invented |
| 12:02.17 | FatPhil | its desperate attempts to remain stateless means that it is slower than swimming through treacle in january |
| 12:02.25 | DocScrutinizer05 | and about repo trust levels |
| 12:02.43 | DocScrutinizer05 | :nod: |
| 12:02.59 | oooaaaooo | what does the H in HAM stand for |
| 12:03.03 | FatPhil | uses fapman quite happily presently. |
| 12:03.04 | DocScrutinizer05 | just don't use apt to do maemo release upgrades |
| 12:03.09 | DocScrutinizer05 | Hildon |
| 12:03.30 | DocScrutinizer05 | FatPhil: fapman is a GUI wrapper around apt |
| 12:03.55 | DocScrutinizer05 | it doesn't know nor care about the nokia idiosyncrasies in repo management |
| 12:03.57 | oooaaaooo | i have app manager but dont see a HAM icon |
| 12:04.03 | Sicelo | uses apt |
| 12:04.12 | DocScrutinizer05 | Hildon "Application Manafer" |
| 12:04.21 | DocScrutinizer05 | s/f/g/ |
| 12:04.56 | DocScrutinizer05 | Sicelo: everybody does. Just not for everything |
| 12:05.19 | Sicelo | well, for everything with me.. |
| 12:05.21 | DocScrutinizer05 | there are a few packages you better use HAM for |
| 12:05.34 | jonwil | And I will create http://wiki.maemo.org/Porting/Closed_Packages for discussing the closed packages and what to do with them (basing the info on some assumptions about the hardware that will be different on the Neo900 vs the N900) |
| 12:05.34 | Sicelo | of couse i never do 'apt-get upgrade -y' :D |
| 12:05.43 | DocScrutinizer05 | like maemo OS and maemoCSSU OS upgrades |
| 12:06.01 | DocScrutinizer05 | or system stuff like bootmenu or kernels |
| 12:06.24 | DocScrutinizer05 | you *can* use apt for that, if you know *exactly* what you're doing |
| 12:07.00 | DocScrutinizer05 | jonwil: sounds good to me |
| 12:09.44 | oooaaaooo | so is HAM the default package manager? |
| 12:10.39 | jonwil | btw, did we say that we wont be using the same Broadcom BCM2048 Bluetooth chip as the N900? I know the WiFi chip wont be a WL1251... |
| 12:11.21 | oooaaaooo | oh its just a front end |
| 12:12.31 | Sicelo | oooaaaooo: default (and only) :P |
| 12:14.12 | jonwil | ok, going to assume we wont be using the same bluetooth chip :) |
| 12:15.15 | oooaaaooo | Sicelo: im trying to follow this http://wiki.maemo.org/Community_SSU . So ill just launch the HAM and look for rootsh and then install backupmenu via dpkg? |
| 12:17.39 | Sicelo | yes. or even backupmenu with HAM as well |
| 12:18.07 | oooaaaooo | Sicelo: oh ... i knew that ! :) |
| 12:19.50 | oooaaaooo | Sicelo: umm how do i search ham? |
| 12:25.19 | oooaaaooo | ok so i go to application manager > download > all then type in "back" and only see microfeed & web2sms |
| 12:25.49 | oooaaaooo | nvm i c rootsh |
| 12:26.03 | oooaaaooo | whatbout sudser? |
| 12:26.20 | Sicelo | i think u wil regret installing that |
| 12:26.35 | oooaaaooo | Sicelo: sudser? |
| 12:26.49 | Sicelo | yes |
| 12:27.04 | oooaaaooo | Sicelo: ummm y? |
| 12:27.40 | oooaaaooo | ok nvm ive downloaded it |
| 12:27.45 | oooaaaooo | rootsh |
| 12:28.11 | Sicelo | maybe i'm the one that made a mistake, but it changed some of my root settings and i could never get them back afterwards |
| 12:29.23 | *** join/#maemo-ssu raccoon_ (user@ghs/raccoon) |
| 12:29.25 | oooaaaooo | Sicelo: ok |
| 12:29.59 | oooaaaooo | so whats the key for " | " (pipe) on the n900 |
| 12:30.11 | oooaaaooo | like if i wanted to | less |
| 12:30.26 | Sicelo | Fn+Ctrl.. then choose from characters |
| 12:30.39 | oooaaaooo | kool |
| 12:30.51 | oooaaaooo | thx' |
| 12:31.24 | Sicelo | you can remap they keeyboard though if you've brave.. and can also add extra characters onto Xterm's 'command bar' if i may call it that |
| 12:31.45 | oooaaaooo | ok so less isnt installed |
| 12:32.02 | oooaaaooo | so isnt man |
| 12:33.37 | Sicelo | ~jr-tools |
| 12:33.37 | infobot | [jrtools] http://wiki.maemo.org/User:Joerg_rw/tools |
| 12:33.56 | Sicelo | oooaaaooo: i guess you're looking for that page ... |
| 12:37.27 | oooaaaooo | Sicelo: oh cool bash |
| 12:39.29 | oooaaaooo | Sicelo: how big's your home |
| 12:42.41 | Sicelo | i think mine is 8GB.. didn't use DocScrutinizer05's way though.. because i wanted to avoid a reflash |
| 12:43.17 | Sicelo | ah, it's 6GB. i once had 8GB though |
| 12:46.58 | oooaaaooo | Sicelo: so that link thats docs way? |
| 12:47.44 | Sicelo | yup.. |
| 12:47.45 | Sicelo | :P |
| 12:48.09 | Sicelo | ~Docscrutinizer |
| 12:48.09 | infobot | i heard docscrutinizer is jOERG, a HW-developer and engineer of Openmoko |
| 12:48.17 | DocScrutinizer05 | hmm? |
| 12:48.33 | Sicelo | don't worry, positive gossip :D |
| 12:48.38 | DocScrutinizer05 | aah |
| 12:51.06 | oooaaaooo | so what exactly is the image being flashed? |
| 12:51.11 | oooaaaooo | is that the cssu? |
| 12:53.42 | Sicelo | no. there's no cssu image. you flash normal PR1.3, then upgrade to CSSU |
| 12:56.30 | oooaaaooo | Sicelo: ok i have 21.2011.38 so thats PR1.3.1? |
| 12:57.31 | oooaaaooo | so i wouldnt be able to just ssh in and do the NEW-HOME size thingy? |
| 12:57.46 | Sicelo | no |
| 12:58.02 | Sicelo | you must re-partition |
| 12:58.22 | Sicelo | there's a detailed wiki page about how you do that (which is what i did) |
| 12:58.55 | oooaaaooo | Sicelo: link? |
| 12:59.28 | Sicelo | ugh.. /me has never been good with bookmarks.. let me look for it |
| 13:00.05 | *** join/#maemo-ssu discopig (~discopig@unaffiliated/discopig) |
| 13:00.24 | oooaaaooo | Sicelo: thanks man |
| 13:01.38 | Sicelo | http://wiki.maemo.org/Repartitioning_the_flash |
| 13:01.39 | oooaaaooo | Sicelo: i wonder if in storage mode gparted picks it up? |
| 13:04.06 | oooaaaooo | ahh |
| 13:04.08 | oooaaaooo | :) |
| 13:07.15 | oooaaaooo | ok so i need to install backup menu |
| 13:07.36 | Sicelo | very handy |
| 13:07.57 | DocScrutinizer05 | you don't regularly need to increase the 2GB $HOME you got |
| 13:08.32 | DocScrutinizer05 | 99.5% of users have their 2GB /home partition at <50% usage |
| 13:08.41 | DocScrutinizer05 | 2GB is a *lot* |
| 13:09.12 | Sicelo | true DocScrutinizer05. i'm only using 1.4GB atm |
| 13:09.32 | chem|st | DocScrutinizer05: $User has no idea how to apt-get |
| 13:10.19 | DocScrutinizer05 | well, apt-get is largely unrelated to /home anyway |
| 13:10.33 | DocScrutinizer05 | unless package is optified |
| 13:10.54 | chem|st | DocScrutinizer05: talking about users installing stuff from testing when on stable |
| 13:11.06 | DocScrutinizer05 | aah |
| 13:11.15 | DocScrutinizer05 | that's not the point. |
| 13:11.22 | oooaaaooo | DocScrutinizer05: im new to martphones(this is my first smartphone...late boomer i know) so im not really sure what i would want to install actually |
| 13:11.46 | oooaaaooo | DocScrutinizer05: maybe a couple of emulators and some roms as well |
| 13:12.00 | DocScrutinizer05 | those will indeed eat lots on /home |
| 13:12.46 | DocScrutinizer05 | read: |
| 13:12.51 | DocScrutinizer05 | ~optification |
| 13:12.51 | infobot | optification is a inventive duct tape workaround to reclaim space in fs root, done due to the fact the systeminit *and* partitioning is FUBAR, http://wiki.maemo.org/Documentation/Maemo_5_Developer_Guide/Packaging,_Deploying_and_Distributing/Installing_under_opt_and_MyDocs, or ""OMG - I wish they looked into FHS and moved /usr to eMMC"", http://www.pathname.com/fhs/pub/fhs-2.3.html#PURPOSE2 bullet1,2 and fhs-2.3.html#PURPOSE16 dot3" |
| 13:12.59 | DocScrutinizer05 | the first link |
| 13:14.19 | DocScrutinizer05 | chem|st: the point is: when you *could* use apt-get, then there's no reason to include anything like nicocam per default, since you as well can distribute the package via .install files or any other convenient means you think is appropriate |
| 13:15.08 | DocScrutinizer05 | "click here to get CSSU cam into cssu-stable" |
| 13:15.41 | DocScrutinizer05 | no new repos needed, no mandatory forced new cameraui in metapackage |
| 13:15.58 | DocScrutinizer05 | fredom of choice |
| 13:16.58 | DocScrutinizer05 | even a "click here to *get rid* of new cameraui and roll back to stock camera in your CSSU-S system" is just as fine |
| 13:17.35 | DocScrutinizer05 | jus definitely no mandatory installation of new bits that are not technically mandatory |
| 13:18.20 | DocScrutinizer05 | easy to grok that philosophy, since it's default policy in all known linux distros |
| 13:18.33 | DocScrutinizer05 | except maemo |
| 13:20.59 | DocScrutinizer05 | technically mandatory bits that can't get uninstalled by CSSU users are: security fixes. essential bug fixes for bugs that would break core system functions. And any stuff that other core system functions DEPEND on the new changed API |
| 13:21.19 | DocScrutinizer05 | nothing of that applies to cameraui |
| 13:21.25 | *** join/#maemo-ssu BCMM (~user@unaffiliated/bcmm) |
| 13:21.44 | DocScrutinizer05 | so it clearly is user's call to decide if they wnat or don't want it on their system |
| 13:22.58 | oooaaaooo | DocScrutinizer05: who is DocScrutinizer51? |
| 13:23.09 | DocScrutinizer05 | my N900 client |
| 13:23.26 | DocScrutinizer05 | N900 == RX51 |
| 13:23.42 | oooaaaooo | o ok |
| 13:24.55 | DocScrutinizer05 | at times there also been a *8 for N810 |
| 13:25.46 | DocScrutinizer05 | oooaaaooo: try /whois <nickname> |
| 13:26.04 | DocScrutinizer05 | and /msg nickserv info <nickname> |
| 13:26.30 | oooaaaooo | cool |
| 13:27.06 | oooaaaooo | DocScrutinizer05: so when i hook up my n900 to my pc in storage mode and it mounts as /495c-a960/dcim, where is that in the N900? |
| 13:27.22 | Sicelo | Camera :D |
| 13:27.40 | DocScrutinizer05 | usually in /home/user/MyDocs/DCIM |
| 13:27.48 | DocScrutinizer05 | or on uSD |
| 13:28.04 | oooaaaooo | DocScrutinizer05: ok lemme see |
| 13:28.11 | DocScrutinizer05 | which is err /media/card/.... iirc |
| 13:28.34 | Sicelo | /media/mmc1 |
| 13:28.38 | DocScrutinizer05 | ok |
| 13:29.19 | Sicelo | the one in MyDocs is called Camera in stock File Manager |
| 13:29.41 | DocScrutinizer05 | yep, it has this insane hiding_the_real_name thing |
| 13:30.15 | oooaaaooo | i just copied backupmenu to /dcim on my pc but it isnt in /home/user/MyDocs/DCIM |
| 13:30.18 | DocScrutinizer05 | which confuses the hell outa noobs and experienced users alike |
| 13:31.01 | DocScrutinizer05 | it's in MyDocs/dcim I guess |
| 13:31.01 | Sicelo | hmm, |
| 13:31.55 | DocScrutinizer05 | you might notice UpperCase/lowercase issues on linux, that you're not used to on windows |
| 13:32.12 | oooaaaooo | im on linux |
| 13:32.43 | DocScrutinizer05 | :nod: thus I said "it's in MyDocs/dcim I guess" |
| 13:32.58 | DocScrutinizer05 | "...not in DCIM" |
| 13:33.51 | DocScrutinizer05 | you usually wouldn't want to copy sth like "backupmenu" to DCIM anyway |
| 13:33.55 | oooaaaooo | /dev/sdb on /media/495C-A960 thats what mount says |
| 13:33.56 | DocScrutinizer05 | it's no photo |
| 13:34.14 | oooaaaooo | DocScrutinizer05: ok lol, parent directory then? |
| 13:34.20 | DocScrutinizer05 | better idea |
| 13:37.36 | DocScrutinizer05 | actually installing .dpkg (which is what I guess you're about to do, for whatever reason) is pretty simple on N900: click on the .dpk URL in webbrowser, it asks you whether to open or download. Do download. Then use filemanager to navigate to the download location which happens to be somewhere on MyDocs, and click the downloaded .dpkg which installs it |
| 13:38.00 | DocScrutinizer05 | no shell needed at all, no dpkg -i cmdline or anything |
| 13:38.18 | Sicelo | DocScrutinizer05 has grown old :p |
| 13:38.48 | *** join/#maemo-ssu LaoLang_cool (~LaoLang_c@112.96.172.87) |
| 13:38.50 | Sicelo | s/.dpkg/.deb/ |
| 13:38.56 | DocScrutinizer05 | yeah, old enough to know $USER |
| 13:39.02 | DocScrutinizer05 | indeed |
| 13:39.06 | DocScrutinizer05 | sorry |
| 13:39.08 | DocScrutinizer05 | .deb |
| 13:39.23 | oooaaaooo | DocScrutinizer05: WAIT SO I CAN INSTALL .DEB FILES USING HAM? |
| 13:39.32 | DocScrutinizer05 | basically yes |
| 13:39.37 | DocScrutinizer05 | via filemanager |
| 13:39.47 | oooaaaooo | DocScrutinizer05: facepalm |
| 13:39.52 | oooaaaooo | DocScrutinizer05: lol |
| 13:40.00 | oooaaaooo | DocScrutinizer05: thanks |
| 13:40.03 | DocScrutinizer05 | yw |
| 13:40.39 | Sicelo | and of course, don't make a habit of installing .debs.. even on a real debian system you're advised to use a 'higher-level' package manager than dpkg |
| 13:40.53 | DocScrutinizer05 | it *should* work even with browser only, but for unclear reasons it fails to install the .deb after downloading it, when you choose "open" |
| 13:40.59 | DocScrutinizer05 | instead of "download" |
| 13:42.01 | oooaaaooo | Sicelo: yeah but i couldnt find backupmenu in HAM |
| 13:42.17 | DocScrutinizer05 | should be there |
| 13:42.24 | DocScrutinizer05 | in extras-devel catalog |
| 13:42.33 | oooaaaooo | catalog...? |
| 13:42.35 | DocScrutinizer05 | and now even in testing, iirc |
| 13:42.39 | oooaaaooo | is that like a repo? |
| 13:42.49 | DocScrutinizer05 | catalog == HAM's term for "repo" |
| 13:43.04 | DocScrutinizer05 | yes |
| 13:43.21 | DocScrutinizer05 | it *is* a repo |
| 13:43.44 | DocScrutinizer05 | in HAM click menu, click "catalogs" |
| 13:43.48 | oooaaaooo | yup |
| 13:43.53 | oooaaaooo | loading it says |
| 13:44.00 | oooaaaooo | or |
| 13:44.02 | oooaaaooo | sorry |
| 13:44.04 | *** join/#maemo-ssu dos1 (~dos@unaffiliated/dos1) |
| 13:44.08 | oooaaaooo | done |
| 13:44.17 | oooaaaooo | 9i see 4 cats |
| 13:44.24 | oooaaaooo | i see 4 catalogs |
| 13:44.43 | DocScrutinizer05 | one of them is "maemo extras" |
| 13:44.44 | oooaaaooo | nokia x2, ovi, maemo |
| 13:44.48 | oooaaaooo | no |
| 13:44.52 | oooaaaooo | maemo.org |
| 13:45.00 | DocScrutinizer05 | another one you may want is "maemo extras-devel" |
| 13:45.25 | oooaaaooo | so where are the catalog details ? |
| 13:45.31 | oooaaaooo | to manually add them |
| 13:46.31 | oooaaaooo | http://talk.maemo.org/showthread.php?t=78584 |
| 13:46.34 | oooaaaooo | is thta it? |
| 13:51.37 | oooaaaooo | lsb_release & cat /etc/*-release doesnt work ... how do i check my distro release ? |
| 14:00.53 | oooaaaooo | ping |
| 14:01.08 | jonwil | There is no concept of "distro release" on Mameo |
| 14:01.56 | jonwil | not in the normal Linux or Debian sense anyway |
| 14:02.00 | oooaaaooo | http://talk.maemo.org/showthread.php?t=78584 |
| 14:02.13 | oooaaaooo | whats this freemantle stuff |
| 14:06.43 | oooaaaooo | ok so i cant install the deb ; dependencies so i should do it via HAM |
| 14:08.57 | oooaaaooo | DocScrutinizer05: so how do i add the repos you mentioned |
| 14:09.35 | oooaaaooo | DocScrutinizer05: nevermind im reading http://wiki.maemo.org/Repository |
| 14:09.49 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 14:09.55 | DocScrutinizer05 | wait |
| 14:10.16 | DocScrutinizer05 | click on the link in jr_tools page |
| 14:10.24 | DocScrutinizer05 | simpler |
| 14:10.41 | DocScrutinizer05 | ~jrtools |
| 14:10.41 | infobot | extra, extra, read all about it, jrtools is http://wiki.maemo.org/User:Joerg_rw/tools |
| 14:11.03 | DocScrutinizer05 | http://wiki.maemo.org/User:Joerg_rw/tools#Additional_Repositories |
| 14:12.41 | oooaaaooo | k so i navigate there on my n900 and just click? |
| 14:13.04 | DocScrutinizer05 | yep, you click on http://joerg.cloud-7.de/repositories.install |
| 14:13.44 | DocScrutinizer05 | it adds 3 catalogs, you will want to delete (or disable at very least) the extras-testing and tools catalogs |
| 14:13.49 | DocScrutinizer05 | in HAM |
| 14:14.01 | oooaaaooo | k |
| 14:15.06 | DocScrutinizer05 | you same time can also rename the display-name to sth more convenient. Right now it's "name = Maemo Extras-DEVEL catalogue (read WARNING!)" |
| 14:15.30 | DocScrutinizer05 | in HAM it will shpow up as "Maemo Extras-DEVEL catalogue (read WARNING!)" |
| 14:15.39 | DocScrutinizer05 | you can edit that name to your liking |
| 14:18.22 | DocScrutinizer05 | just never do an "update all" in HAM when any of those catalogs is enabled. Also never do an apt-get upgrade |
| 14:18.46 | oooaaaooo | ok so no updates when these extras are enabled |
| 14:18.56 | oooaaaooo | i thought apt-get was to never be used |
| 14:19.34 | DocScrutinizer05 | you can use apt-get install, for stuff like less or binutils-gnu or the like. |
| 14:19.41 | oooaaaooo | ok |
| 14:20.10 | oooaaaooo | out of curiosity, is aptitude available? |
| 14:20.16 | DocScrutinizer05 | you just shouldn't use it for anything that sounds like maemo specific |
| 14:20.23 | oooaaaooo | oh ok |
| 14:20.23 | DocScrutinizer05 | afaik it is |
| 14:20.31 | oooaaaooo | ok |
| 14:21.07 | oooaaaooo | hmmm |
| 14:21.17 | oooaaaooo | that update thing |
| 14:21.27 | oooaaaooo | i think my ham is set to auto update |
| 14:21.36 | DocScrutinizer05 | as a rule of thumb: when it's a "normal" debian package then you can use apt-get install to both install as well as update the package |
| 14:21.47 | oooaaaooo | ok |
| 14:22.40 | DocScrutinizer05 | yes, ham usually checks once per day for upgrades available. But it asks you if they should get installed. Just answer "No! Ignore!" |
| 14:22.40 | oooaaaooo | where HAM says checking for updates |
| 14:22.48 | oooaaaooo | ok |
| 14:22.55 | oooaaaooo | this is everytime i load HAM |
| 14:23.00 | oooaaaooo | its annoying |
| 14:23.16 | DocScrutinizer05 | that's why you disable the extras-devel catalog |
| 14:23.28 | DocScrutinizer05 | after you installed the package you're interested in |
| 14:23.57 | oooaaaooo | i dont know if i was successful in adding the repos |
| 14:24.02 | oooaaaooo | oh wait |
| 14:24.10 | oooaaaooo | i c it |
| 14:24.13 | oooaaaooo | them |
| 14:24.30 | DocScrutinizer05 | with extras-devel disabled, HAM won't upgrade any packages from extras-devel and only show upgrades available in the "good" repos |
| 14:25.22 | DocScrutinizer05 | all upgrades from "good" repos are safe to install |
| 14:25.32 | DocScrutinizer05 | actually recommended to install |
| 14:26.04 | DocScrutinizer05 | avoid upgrades of any packages that come from "non-good" repos like extras-devel |
| 14:26.17 | oooaaaooo | ok so enable devel extras , look for backup menu then can disable |
| 14:26.25 | DocScrutinizer05 | this is why you disable the extras-devel repo |
| 14:26.31 | oooaaaooo | ok |
| 14:26.33 | DocScrutinizer05 | yes, exactly |
| 14:26.56 | oooaaaooo | umm is there a quick and easy way for stock n900 to be tethered to my pc's eth0 |
| 14:27.00 | freemangordon | DocScrutinizer05: hmm, why this is not in extras? |
| 14:27.02 | oooaaaooo | via usb |
| 14:27.13 | DocScrutinizer05 | dunno, is it? |
| 14:27.25 | freemangordon | what is tha package name? |
| 14:27.30 | DocScrutinizer05 | afaik it has unmet dependencies |
| 14:27.58 | DocScrutinizer05 | e.g to i2ctools |
| 14:28.16 | DocScrutinizer05 | for charging |
| 14:28.27 | freemangordon | ah, I see |
| 14:28.49 | DocScrutinizer05 | we might promote i2ctools to extras (hidden) |
| 14:29.03 | DocScrutinizer05 | to allow BM to promote to extras |
| 14:29.55 | DocScrutinizer05 | aiui HAM wil not show i2ctools when it's not a user package |
| 14:30.16 | DocScrutinizer05 | I don't care a lot about fapman users accidentally installing i2ctools from extras repo |
| 14:32.54 | oooaaaooo | yeah i still dont see backupmenu.... |
| 14:33.24 | DocScrutinizer05 | oooaaaooo: since that's depending a lot on the particular OS you run on your PC, there's no good general instruction for it |
| 14:34.05 | DocScrutinizer05 | oooaaaooo: wait a minute, I'll check that friggin BM thing, it lately seen some mess-up |
| 14:34.14 | oooaaaooo | DocScrutinizer05: ok, thnx |
| 14:36.36 | DocScrutinizer05 | a scray sight what HAM thinks about availability of stock core repos |
| 14:37.38 | freemangordon | Pali: http://pastebin.com/i5CJWWzq :D |
| 14:38.06 | DocScrutinizer05 | "checking updates" takes ages (like 5 min) after enabling extras-devel |
| 14:38.25 | oooaaaooo | DocScrutinizer05: i kept stopping the updates...is that why? |
| 14:38.38 | DocScrutinizer05 | yes |
| 14:38.48 | freemangordon | Pali: both drivers loaded, with clock framework, lets see if back camera works |
| 14:38.49 | DocScrutinizer05 | you mustn't stop updates |
| 14:39.10 | oooaaaooo | DocScrutinizer05: ok so all i did was add the repo addresses when i clicked on that link |
| 14:39.24 | DocScrutinizer05 | yes |
| 14:40.30 | freemangordon | Pali: - entity 16: et8ek8 3-003e (1 pad, 1 link) |
| 14:40.32 | freemangordon | type V4L2 subdev subtype Unknown flags 0 |
| 14:40.34 | freemangordon | device node name /dev/v4l-subdev8 |
| 14:40.36 | freemangordon | pad0: Source |
| 14:40.37 | DocScrutinizer05 | now go to HAM, and wait for it to finish "checking for updates" |
| 14:40.38 | freemangordon | [fmt:SGRBG10_DPCM8/2592x1968] |
| 14:40.40 | freemangordon | -> "OMAP3 ISP CCP2":0 [] |
| 14:40.51 | oooaaaooo | DocScrutinizer05: yup |
| 14:41.03 | oooaaaooo | DocScrutinizer05: doing it now |
| 14:41.04 | DocScrutinizer05 | will take ages |
| 14:41.30 | oooaaaooo | DocScrutinizer05: ok well ill just have to continue it tomorrow. THanks alot for the help |
| 14:41.39 | DocScrutinizer05 | after that, you see "backupmenu" under "install new" |
| 14:41.52 | oooaaaooo | whats install new |
| 14:42.00 | oooaaaooo | is that an icon |
| 14:42.07 | DocScrutinizer05 | backupmenu 1.2.0-2 |
| 14:42.25 | oooaaaooo | ok, i should be able to search it via "all" |
| 14:42.27 | oooaaaooo | right |
| 14:42.29 | DocScrutinizer05 | it's one of the three choices in HAM main screen |
| 14:42.47 | DocScrutinizer05 | actually the middle one |
| 14:43.07 | DocScrutinizer05 | right side is "updates", left side is "installed apps" |
| 14:43.20 | DocScrutinizer05 | middle is "install new" |
| 14:43.50 | DocScrutinizer05 | or "download new" or whatever |
| 14:44.03 | oooaaaooo | ok will figure it out tomorrow |
| 14:44.06 | oooaaaooo | im off to bed |
| 14:44.09 | oooaaaooo | thanks alot guys |
| 14:44.32 | DocScrutinizer05 | in German it's "Deinstallieren" "Herunterladen" "Aktualisieren" |
| 14:44.46 | oooaaaooo | (y) |
| 14:45.34 | DocScrutinizer05 | pick the middle one, go to menu, click "update" (upper left menu item), then search for "backupmenu" |
| 14:45.46 | freemangordon | Pali: ok, back camera is green too, seems it is something in omap3isp :( |
| 14:46.09 | Pali | ah :-( |
| 14:46.57 | DocScrutinizer05 | cam drivers are icky to maintain |
| 14:47.07 | Pali | did you tried to contact sakari? maybe he will know where is problem... |
| 14:47.15 | DocScrutinizer05 | N900 is special since it has that mux |
| 14:47.51 | freemangordon | DocScrutinizer05: I know |
| 14:48.03 | DocScrutinizer05 | k |
| 14:48.10 | freemangordon | DocScrutinizer05: and the camera drivers code is just fine |
| 14:48.44 | freemangordon | Pali: no, I wanted to first try main camera too, to see if the problem is in camera driver or in omap3isp driver |
| 14:49.13 | freemangordon | DocScrutinizer05: funny thing is that upstream kernel does not support more than one camera connetced to one and the same clk :D |
| 14:49.28 | DocScrutinizer05 | freemangordon: make sure you load/initialize any kernel cam driver module *after* you checked and switched mux |
| 14:49.45 | freemangordon | DocScrutinizer05: yep, init code is fine |
| 14:49.53 | DocScrutinizer05 | that's the systematic problem with upstream drivers I guess |
| 14:49.59 | freemangordon | :nod: |
| 14:50.29 | freemangordon | Pali: I will contact sakari |
| 14:50.39 | DocScrutinizer05 | they do init on mod load, and mod load is done only once. Then you switch mux and everything goes boom instantly |
| 14:50.43 | freemangordon | and we'll have to upstream a couple of patches :D |
| 14:51.10 | Pali | freemangordon: CC mail to me |
| 14:51.13 | freemangordon | DocScrutinizer05: no,no, drivers are loaded just fine, they talk with camera modules, etc |
| 14:51.36 | freemangordon | DocScrutinizer05: it is just that all I see is a green square in mplayer |
| 14:51.41 | DocScrutinizer05 | and no mux xswitching after they got loaded and did init? |
| 14:51.47 | freemangordon | rectangle even |
| 14:51.58 | freemangordon | xshutdown is controlled |
| 14:52.05 | freemangordon | so there is mux switch |
| 14:52.15 | freemangordon | it is done in board code, via callbacks |
| 14:52.27 | DocScrutinizer05 | as soon as you switch the mux, both cam-modules/chips are no more initialized |
| 14:53.12 | DocScrutinizer05 | you first need to switch mux, then power up the cam and load the driver which runs init |
| 14:53.37 | freemangordon | DocScrutinizer05: yes, board init code disables reset |
| 14:53.39 | DocScrutinizer05 | Nokia drivers are taking care of all that I guess |
| 14:53.55 | DocScrutinizer05 | disable reset - niiiiice |
| 14:53.59 | DocScrutinizer05 | :-S |
| 14:54.01 | freemangordon | it seems upstream does it as well |
| 14:54.13 | freemangordon | DocScrutinizer05: why? |
| 14:54.52 | freemangordon | disable reset->xshutdown up->power up->init |
| 14:54.52 | DocScrutinizer05 | I'd think you need to reste the camchip after switching mux, to get it into shape again |
| 14:55.11 | DocScrutinizer05 | then redo chip init which usually is done in driver-load init |
| 14:55.39 | freemangordon | before streaming is enabled, camera is fed with the needed values for resolution, etc |
| 14:55.45 | freemangordon | I see no problem on that part |
| 14:56.06 | freemangordon | and because both camers behave exactly in the same way, I bet it is isp code to blame |
| 14:56.22 | DocScrutinizer05 | when cam is powered and you switch mux, the chip is in undefined state and never recovers from that until you powercycle and reset it |
| 14:56.28 | freemangordon | I already found (and fixed) a bug in it. It has naver been tested on 3430 only on 3630 |
| 14:56.45 | freemangordon | DocScrutinizer05: this is done, np there |
| 14:56.52 | DocScrutinizer05 | ok then |
| 14:57.00 | DocScrutinizer05 | I never looked into the drivers |
| 14:57.19 | DocScrutinizer05 | just thought I reiterate about the basics |
| 14:57.20 | freemangordon | it is a can of worms :( with really bad design IMO |
| 14:57.24 | freemangordon | ok |
| 14:58.37 | freemangordon | (as I already explained) you can't have more than one device connected to cam_xclka (or cam_xclkb), neither you can use xshutdown gpio for anything alse but xshutdown |
| 15:00.13 | freemangordon | like mux for example :) |
| 15:00.20 | DocScrutinizer05 | yeah |
| 15:00.27 | DocScrutinizer05 | thought as much |
| 15:00.37 | DocScrutinizer05 | incompatible design schemes |
| 15:01.07 | DocScrutinizer05 | the probelm with a "one kernel for all" approach |
| 15:01.49 | DocScrutinizer05 | and in the end what killed power management on all "strictly upstream only" embedded projects |
| 15:02.23 | freemangordon | :nod: |
| 15:06.10 | DocScrutinizer05 | you say "we need to postpone init of B after wit had a chance to switch A". Upstream says "you can't do that! layering violation" |
| 15:06.31 | DocScrutinizer05 | s/wit/we/ |
| 15:07.31 | DocScrutinizer05 | in the end the kernel devels claim that real hardware doesn't meet the requirements ;-P |
| 15:07.50 | freemangordon | yeah :D |
| 15:08.36 | *** join/#maemo-ssu Martix (~martix@123.225.broadband2.iol.cz) |
| 15:09.50 | DocScrutinizer05 | "the CPU has to be up before you can run init for the power regulators! there's no way to do it any other way, our software paradigms allow any violations of this dependency" |
| 15:10.18 | DocScrutinizer05 | dont* allow |
| 15:10.57 | DocScrutinizer05 | or make "CPU " rather "I2C" ;-P |
| 15:11.33 | freemangordon | http://www.kasperskyasia.com/images/alliances/xplogo.gif :P |
| 15:11.37 | DocScrutinizer05 | "you can't run init of the PSU at this time, we haven't initialized I2C yet!" |
| 15:11.59 | Sicelo | why are we doing camera drivers btw? |
| 15:12.08 | freemangordon | Sicelo: linux 3.10 |
| 15:12.14 | Sicelo | aha |
| 15:12.15 | DocScrutinizer05 | "you can't initialize I2C unless you powered it up!" |
| 15:12.17 | freemangordon | and up |
| 15:12.50 | DocScrutinizer05 | "you can't power up I2C before you intitalized PSU!" |
| 15:13.18 | Sicelo | i saw they are forcing also to have N900 on DT.. how much work is that? |
| 15:13.22 | DocScrutinizer05 | always *a lot* of fun to discuss this stuff with kerenl devs |
| 15:13.52 | freemangordon | Sicelo: most probably wont fly, see the cameras example ^^^ |
| 15:14.06 | Sicelo | dang |
| 15:14.35 | DocScrutinizer05 | there's a reason Nokia came up with a maemo-N900 kernel, and not simply using upstream |
| 15:15.03 | Sicelo | sebastian reichel was also working on the modem, and was also told DT. |
| 15:15.31 | DocScrutinizer05 | seen that several times now - in the end some mandatory tweaks/hacks to make your platform fly will never get accepted upstream |
| 15:15.40 | freemangordon | modem *might* be possible |
| 15:16.17 | Sicelo | he did sound optimistic on the ML, yes |
| 15:16.56 | DocScrutinizer05 | evetually you go "screw dat! I fork!" |
| 15:17.17 | *** join/#maemo-ssu X-Fade (~xfade@d5152FFD8.static.telenet.be) |
| 15:17.18 | freemangordon | wonders what we will do with 170mA current on idle with 3.10 |
| 15:17.30 | DocScrutinizer05 | ...and have a chance to end with a *working* kernel, though not upstream |
| 15:17.53 | DocScrutinizer05 | freemangordon: live with it? |
| 15:17.58 | freemangordon | hehe |
| 15:18.12 | freemangordon | no way, I'll just stay with kp |
| 15:19.21 | DocScrutinizer05 | dang, thursday already |
| 15:19.55 | Sicelo | freemangordon: how is kp able to do it, and what stops that woking on 3.x? |
| 15:20.05 | DocScrutinizer05 | see above |
| 15:20.27 | DocScrutinizer05 | it's Nokia's kernel patches that do the magic |
| 15:20.35 | DocScrutinizer05 | and they will never go mailine |
| 15:20.41 | DocScrutinizer05 | main line even |
| 15:20.42 | *** join/#maemo-ssu BCMM (~user@unaffiliated/bcmm) |
| 15:20.49 | DocScrutinizer05 | aka upstream |
| 15:20.57 | Sicelo | but they are not closed-source? |
| 15:21.03 | DocScrutinizer05 | not afaik |
| 15:21.08 | DocScrutinizer05 | just "dirty" |
| 15:21.31 | DocScrutinizer05 | kernel has no closed blobs |
| 15:21.58 | DocScrutinizer05 | it however may have some missing parts in maemo that are emulated with userland blob.so |
| 15:22.09 | freemangordon | Sicelo: nothing closed-source in the kernel, patches just doesn't "fit" in what upstream maintainters think is the right kernel structure |
| 15:22.10 | DocScrutinizer05 | powervr |
| 15:22.18 | *** join/#maemo-ssu X-Fade (~xfade@d5152FFD8.static.telenet.be) |
| 15:22.45 | DocScrutinizer05 | bme |
| 15:22.57 | DocScrutinizer05 | a lot of stuff actually |
| 15:23.06 | freemangordon | cmt |
| 15:23.13 | DocScrutinizer05 | :nod: |
| 15:23.16 | freemangordon | gps |
| 15:23.19 | DocScrutinizer05 | :nod: |
| 15:23.29 | freemangordon | I think that' s all |
| 15:23.45 | DocScrutinizer05 | you *can* do all this in kernel space, but you lose the magic done in Nokia's blobs |
| 15:24.18 | freemangordon | not for BME ;), I think we really don;t need the magic coded there :D |
| 15:24.43 | DocScrutinizer05 | for the other open parts, they are just not what upstream is willing to accept since it can't go into their rule-set for how kernel source is structured |
| 15:25.04 | DocScrutinizer05 | freemangordon: :nod: |
| 15:25.35 | Sicelo | structuring the code to fit upstream's expectations.. too hard? |
| 15:27.12 | DocScrutinizer05 | simple madeup example: Nokia: "here's our awesome acelerometer driver, saves lots of energy". Upstream: "erryeah, but it uses IRQ which are not yet initialized at that moment when the accel gets loaded, so you can't do that. We have a poll-only policy for this class of drvers, for exactly that reason" |
| 15:27.56 | Sicelo | ok |
| 15:28.05 | freemangordon | Sicelo: see cameras example ^^^ :) |
| 15:28.15 | freemangordon | it is not hard, but impossible |
| 15:28.24 | DocScrutinizer05 | hehe, exactly |
| 15:29.17 | Sicelo | i understand. |
| 15:29.44 | DocScrutinizer05 | put simple: when you build a kernel for your embedded, you want it to fit your hardware, not match any upstream policy |
| 15:30.00 | Sicelo | NeoN900 going to be too expensive for me. might as well start getting used to S40 :p |
| 15:30.34 | Sicelo | s/oN9/o9/ |
| 15:31.12 | freemangordon | the only option we have aiui is to upstream whatever possible and to keep the other stuff in a separate tree |
| 15:31.24 | DocScrutinizer05 | "Dear Mr council" SPAAAAM! |
| 15:49.11 | *** join/#maemo-ssu Martix (~martix@123.225.broadband2.iol.cz) |
| 15:55.37 | kerio | DocScrutinizer05: "poll-only policy" where? |
| 15:57.14 | Sicelo | it was an example, iirc |
| 15:59.27 | DocScrutinizer05 | kerio: read again! "made-up example" |
| 16:50.14 | *** join/#maemo-ssu NIN101 (~NIN@p5DD28D57.dip0.t-ipconnect.de) |
| 16:53.33 | *** join/#maemo-ssu Martix (~martix@123.225.broadband2.iol.cz) |
| 16:56.03 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 17:11.32 | *** join/#maemo-ssu arcean (~arcean@aacr13.neoplus.adsl.tpnet.pl) |
| 17:33.49 | *** join/#maemo-ssu arcean_ (~arcean@aada194.neoplus.adsl.tpnet.pl) |
| 18:52.03 | *** join/#maemo-ssu M4rtinK (~M4rtinK@46.227.13.46) |
| 18:58.00 | *** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali) |
| 19:49.55 | freemangordon | Pali: oh, who wrote camera board code? |
| 19:50.27 | Pali | I think sakari |
| 19:51.03 | freemangordon | not only there is an unprotected(by the mutex) variable in rx51_camera_set_xshutdown, but also there is a nice race while the modules get loaded :( |
| 19:51.29 | freemangordon | and also when and used later |
| 19:51.41 | freemangordon | s/when and/when/ |
| 19:52.42 | freemangordon | hmm, yeah, sakari, according to the file header |
| 19:56.03 | freemangordon | Pali: will push the changes and will mail him |
| 19:56.17 | Pali | ok |
| 20:01.40 | DocScrutinizer05 | (race) \o/ |
| 20:02.23 | freemangordon | DocScrutinizer05: yeah, I wonder if we have the same in 2.6.28 and if not, how did Nokia manage to fix that |
| 20:02.53 | *** join/#maemo-ssu arcean (~arcean@aafr52.neoplus.adsl.tpnet.pl) |
| 20:02.55 | ShadowJK | Is this why cssucam is crashy |
| 20:03.04 | DocScrutinizer05 | some races hardly ever happen, thanks to properties of scheduler. Others do, every single time |
| 20:03.36 | freemangordon | DocScrutinizer05: kernel tries to power front camera up, while the back camera lens is still powered on. Ofc this is not possible and board gode returns EBUSY |
| 20:03.41 | DocScrutinizer05 | e.g the race between detecting VBUS and D+/- short |
| 20:04.32 | ShadowJK | remembers a bug report about back camera lens getting abrubtly unpowered instead of first returning lens to neutral position |
| 20:05.05 | freemangordon | this could happen while the devices are being initialies or when they are enumerated from userspace. this could be a bug in media-ctl as well, but I am almost sure kernel should have a tools to prevent such conditions |
| 20:05.06 | ShadowJK | in a quiet environment, you can hear the klonk of the lens violently slamming to its unpowered position |
| 20:05.39 | freemangordon | ShadowJK: this is linux 3.10 |
| 20:06.10 | ShadowJK | hey, does fstrim work in 3.10? |
| 20:06.25 | freemangordon | ~fstrim |
| 20:06.37 | freemangordon | wtf is that? |
| 20:08.03 | Pali | maybe trim command for filesystems? |
| 20:08.11 | ShadowJK | manually issues TRIM of free space to storage device.. Some emmc/sd cards support it, if driver stack/kernel also supports it |
| 20:08.17 | ShadowJK | Yes what Pali said |
| 20:09.00 | Pali | I think that vfat in linux kernel has trim support |
| 20:09.26 | Pali | so it can be usefull for maemo mydocs |
| 20:12.57 | DocScrutinizer05 | ShadowJK: heard that KLONK several times |
| 20:14.15 | DocScrutinizer05 | pretty nasty |
| 20:14.26 | Pali | what it cause? |
| 20:14.43 | Pali | autofocus? |
| 20:14.53 | DocScrutinizer05 | err, prolla yhwat ShadowJK said? |
| 20:15.05 | DocScrutinizer05 | yep, related to autofocus |
| 20:17.00 | Pali | autofocus ad5820: Voice coil motor controller |
| 20:17.04 | DocScrutinizer05 | I also still wonder if nicocam stopped displaying random bogus distance readings for autofocus. Rather - disatnce readings that are ~factor two apart from each other for shoot upwards vs shoot the floor |
| 20:17.29 | DocScrutinizer05 | Pali: exactly this one. A nasty power hog as well |
| 20:18.36 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 20:18.44 | DocScrutinizer05 | *I* wonder *who* and *how* *does* autofocus |
| 20:18.53 | ShadowJK | heh, gravity pulling the lens? |
| 20:18.59 | DocScrutinizer05 | ShadowJK: exactly |
| 20:19.13 | DocScrutinizer05 | changing distance of 50cm to 30 or 90 |
| 20:19.43 | Pali | DocScrutinizer05: I bet that proprietary nokia daemon omap3camd |
| 20:19.53 | DocScrutinizer05 | yup, likely |
| 20:19.58 | freemangordon | I bet on that one too |
| 20:20.32 | DocScrutinizer05 | must have some nasty math |
| 20:20.44 | DocScrutinizer05 | to check if focused or not |
| 20:20.54 | DocScrutinizer05 | FFT |
| 20:21.30 | Pali | why nokia not used opencv or libv4l2 libraries for that? |
| 20:21.38 | DocScrutinizer05 | dunno |
| 20:21.42 | Pali | but rather created own daemon? |
| 20:21.47 | Pali | which we cannot fix? |
| 20:22.13 | DocScrutinizer05 | there's for sure more "magic", like compensating for aberrations of cam, etc |
| 20:22.35 | Pali | these algorithms are well-known and already implemented in open source libraries for image processing... |
| 20:22.51 | DocScrutinizer05 | that knowhow comes by courtesy cam module manuf, and is proprietary |
| 20:22.53 | Sicelo | were they good at that time? |
| 20:23.25 | kerio | isn't fcam free? |
| 20:23.42 | DocScrutinizer05 | I think it is |
| 20:23.58 | Pali | http://fcam.garage.maemo.org/ |
| 20:24.28 | DocScrutinizer05 | but afaik fcam only adding a few calls they missed to implement into stock driver |
| 20:25.01 | freemangordon | Pali: pushed |
| 20:25.06 | Pali | ok |
| 20:26.39 | *** join/#maemo-ssu LauRoman (~LauRoman@5-14-93-219.residential.rdsnet.ro) |
| 20:27.52 | Pali | freemangordon: can you backport also commit a354177f058541b7212230feb2c0da7c464e9b9d? |
| 20:28.08 | Pali | so there will not be problem after migrating to 3.11 (or 3.12)? |
| 20:29.07 | freemangordon | ok, but I'll first email sakari about the grass I have when taking pictures :D |
| 20:29.20 | Pali | ok :-) |
| 20:29.25 | Pali | CC mail to me |
| 20:29.30 | freemangordon | ัััะต |
| 20:29.34 | freemangordon | sure |
| 20:36.42 | *** join/#maemo-ssu xes (~xes@unaffiliated/xes) |
| 20:56.00 | freemangordon | Pali: hmm, wonder what subject to use for the mail |
| 20:56.29 | Pali | help with n900 camera :-) |
| 20:56.33 | freemangordon | ok :) |
| 20:57.18 | freemangordon | Pali: are you sure? |
| 21:00.02 | freemangordon | Pali: sent |
| 21:00.19 | Pali | ok |
| 21:02.19 | *** join/#maemo-ssu g3kk3r (torrkull@lehtori.cc.tut.fi) |
| 21:39.29 | *** join/#maemo-ssu M4rtinK (~M4rtinK@46.227.13.46) |
| 22:39.14 | *** join/#maemo-ssu nox- (noident@freebsd/developer/nox) |
| 22:54.52 | *** join/#maemo-ssu dos11 (~dos@unaffiliated/dos1) |
