| 00:22.28 | *** join/#maemo-ssu dafox (~dafox@ip3e8380c8.speed.planet.nl) |
| 01:08.02 | *** join/#maemo-ssu ruskie (~ruskie@sourcemage/mage/ruskie) |
| 02:03.44 | *** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au) |
| 02:14.41 | jonwil | hi |
| 02:52.41 | *** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn) |
| 03:54.41 | *** join/#maemo-ssu jonwil_ (~jonwil@27-33-137-199.static.tpgi.com.au) |
| 03:56.24 | *** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn) |
| 04:38.15 | *** join/#maemo-ssu piscodig (~discopig@modemcable076.197-131-66.mc.videotron.ca) |
| 06:13.11 | *** join/#maemo-ssu LauRoman (~LauRoman@5-14-93-219.residential.rdsnet.ro) |
| 06:49.32 | *** join/#maemo-ssu discopig (~discopig@unaffiliated/discopig) |
| 07:06.41 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 07:46.46 | *** join/#maemo-ssu amizraa (~amizraa@gateway/tor-sasl/amizraa) |
| 07:56.33 | *** join/#maemo-ssu ruskie (ruskie@sourcemage/mage/ruskie) |
| 07:59.35 | *** join/#maemo-ssu _rd (~rd@p5088FD88.dip0.t-ipconnect.de) |
| 08:12.43 | *** join/#maemo-ssu NIN101 (~NIN@p5DD2B8E4.dip0.t-ipconnect.de) |
| 08:14.46 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 08:24.49 | *** join/#maemo-ssu dafox (~dafox@ip3e8380c8.speed.planet.nl) |
| 08:38.38 | *** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au) |
| 08:39.21 | jonwil | hi |
| 08:53.03 | *** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali) |
| 08:53.45 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 09:22.59 | *** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali) |
| 09:25.55 | merlin1991 | jonwil: I'll be around starting tomorrow, now I gotta drive back from amsterdam to vienna .D |
| 09:26.54 | *** join/#maemo-ssu mkaindl (~mkaindl@ama-dablam.markus-kaindl.de) |
| 09:29.12 | jonwil | Pali was able to help me with the thing I needed :) |
| 09:55.39 | *** join/#maemo-ssu bsdmaniak (~bsdmaniak@std93-20-88-120-139-80.fbx.proxad.net) |
| 10:11.33 | Pali | merlin1991: https://gitorious.org/community-ssu/qt-x11-maemo/merge_requests/5#comment_99873 |
| 10:25.59 | *** join/#maemo-ssu discopig (~discopig@2001:5c0:1400:a::31d) |
| 10:30.15 | *** join/#maemo-ssu BCMM (~BCMM@unaffiliated/bcmm) |
| 10:30.57 | *** join/#maemo-ssu Guest29407 (~discopig@unaffiliated/discopig) |
| 10:32.11 | *** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu) |
| 11:10.45 | *** join/#maemo-ssu arcean (~arcean@aaeq134.neoplus.adsl.tpnet.pl) |
| 11:10.53 | *** part/#maemo-ssu mkaindl (~mkaindl@ama-dablam.markus-kaindl.de) |
| 11:12.42 | *** join/#maemo-ssu mkaindl (~mkaindl@ama-dablam.markus-kaindl.de) |
| 11:28.51 | *** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali) |
| 11:34.32 | freemangordon | Pali: https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/d6b8c92c7544cc5559aaa7edcc48d07a7730d9a2 |
| 11:34.51 | freemangordon | and https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/9b0955924ff3ad171dc66b058209fc08481feb77 |
| 11:35.01 | freemangordon | and https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/26f5416802082147e0029e57bbbf9630989ad54a |
| 11:35.03 | freemangordon | :) |
| 11:35.25 | Pali | ok :-) |
| 11:35.39 | freemangordon | I still don;t have cameras or ssi working |
| 11:35.52 | freemangordon | but at least MUX config should be 99% correct |
| 11:36.16 | freemangordon | i dumped mux config from /dev/mem on 2.6.28 and compared with 3.10 |
| 11:39.42 | *** join/#maemo-ssu arcean_ (~arcean@aaeq134.neoplus.adsl.tpnet.pl) |
| 11:46.29 | Pali | so how was Skry able to use modem and send sms? |
| 11:46.39 | freemangordon | NFC :( |
| 11:47.11 | freemangordon | maybe we should ask him to boot 3.10 with alarm and try |
| 11:48.36 | freemangordon | maybe we miss somtething from the configuration (rx51_defconfig), dunno |
| 11:49.48 | Pali | freemangordon: can you check if you have with last my 3.10 tree (looks like you already merged all my changes) /dev/hw* device? |
| 11:50.15 | freemangordon | compiling it now :) |
| 11:50.21 | Pali | some hw random device (do not know correct name under maemo) |
| 11:50.22 | Pali | ok |
| 11:50.31 | freemangordon | hwrng? |
| 11:50.39 | Pali | yes |
| 11:51.17 | Pali | and check if it working? (use cat and see if that chars are random :D) |
| 11:51.17 | freemangordon | BTW git managed somehow to silently fail to merge rx51_defconfig, besides enabling l2auxsave/resore are there any other changes? |
| 11:51.45 | freemangordon | Pali: ^^^ |
| 11:52.12 | Pali | in rx51_defconfig there are no other changes |
| 11:52.15 | freemangordon | ok |
| 11:52.31 | freemangordon | I wonder why it didn;t tell me anything :() |
| 11:52.37 | freemangordon | :( |
| 11:52.53 | freemangordon | I did a simple pull with rebase |
| 11:53.55 | Pali | freemangordon: maybe there is problem also in uboot? |
| 11:53.56 | Pali | http://git.denx.de/?p=u-boot.git;a=blob_plain;f=board/nokia/rx51/rx51.h;hb=HEAD |
| 11:55.02 | freemangordon | you can bet on that :) |
| 11:55.16 | *** join/#maemo-ssu dos1 (~dos@unaffiliated/dos1) |
| 11:55.51 | freemangordon | though... maybe my config is not 100% correct |
| 11:56.05 | freemangordon | as u-boot doesn;t bring problems with 2.6.28 |
| 11:56.30 | Pali | because 2.6.28 kernel setting anything... |
| 11:57.29 | Pali | freemangordon: I pushed that 3 commits into v3.10-n900 tree |
| 11:58.00 | *** join/#maemo-ssu arcean (~arcean@aaeq134.neoplus.adsl.tpnet.pl) |
| 12:18.36 | Pali | merlin1991, freemangordon: what about creating mirror of community-ssu git repos on github? |
| 12:18.57 | freemangordon | Pali: merlin1991 said he'll do it aiui |
| 12:19.05 | Pali | does anybody have github account? |
| 12:19.10 | Pali | ok |
| 12:19.16 | freemangordon | me :) |
| 12:19.32 | Pali | ok |
| 12:25.23 | freemangordon | Pali: no hwrng in /dev |
| 12:26.27 | Pali | no device with name *hw*? |
| 12:26.34 | freemangordon | no |
| 12:26.45 | Pali | then I need syslog/dmesg output |
| 12:28.32 | freemangordon | Pali: http://pastebin.com/xb5QBiqW |
| 12:29.41 | freemangordon | hmm, will powercycle the device and will retry |
| 12:31.22 | freemangordon | hmm, got kernel opps |
| 12:31.39 | freemangordon | oops even |
| 12:31.59 | *** join/#maemo-ssu discopig (~discopig@modemcable076.197-131-66.mc.videotron.ca) |
| 12:32.36 | freemangordon | it is in et8ek8 :D |
| 12:33.19 | *** join/#maemo-ssu Guest65277 (~discopig@unaffiliated/discopig) |
| 12:35.18 | freemangordon | 18.197540] [<bf2cd164>] (et8ek8_power_on+0x50/0x124 [et8ek8]) from [<bf2cd914>] (et8ek8_registered+0x78/0x4c4 [et8ek8]) |
| 12:35.56 | freemangordon | WTF?!? |
| 12:36.11 | freemangordon | et8ek8.c is not pulled too |
| 12:36.13 | freemangordon | shit |
| 12:38.08 | freemangordon | Pali: seems my build is broken, wait until I rebuild and test again |
| 12:47.16 | *** part/#maemo-ssu mkaindl (~mkaindl@ama-dablam.markus-kaindl.de) |
| 12:49.28 | Pali | freemangordon: http://pastebin.com/7Khu6s95 |
| 12:49.34 | Pali | you are missing one commit |
| 12:50.21 | Pali | $ git cherry-pick 081723f1d5cc72232239dee7d77049456440377b |
| 12:50.23 | Pali | freemangordon run ^^^ |
| 12:50.50 | freemangordon | Pali: I am rebasing now |
| 12:50.57 | freemangordon | again :) |
| 13:05.03 | Pali | freemangordon: you still missing that one commit ^^^^^^ |
| 13:05.23 | freemangordon | https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/d5b97aa802ff650878c059dda8ddba31433e01f6 |
| 13:05.46 | Pali | freemangordon: revert that commit |
| 13:05.58 | Pali | name must be "omap3-rom-rng" |
| 13:06.15 | freemangordon | Pali: see your pastebin :) |
| 13:06.25 | Pali | freemangordon: btw, I included all your commits into v3.10-n900 branch |
| 13:06.41 | freemangordon | I know, just merged |
| 13:06.44 | Pali | freemangordon: see title of that pastebin :-) |
| 13:07.12 | freemangordon | Pali: oh :) |
| 13:07.37 | Pali | freemangordon: I think you can now remove your modified branch |
| 13:07.52 | Pali | because it has all and same commits as which are in global one |
| 13:08.02 | freemangordon | Pali: will do and will colne again |
| 13:08.05 | freemangordon | *clone |
| 13:08.47 | freemangordon | Pali: hmm, https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/081723f1d5cc72232239dee7d77049456440377b |
| 13:09.04 | Pali | yes |
| 13:09.16 | freemangordon | but it was in the tree after the merge |
| 13:09.22 | Pali | this is right, name should be "omap3-rom-rng" |
| 13:09.41 | Pali | yes it was in tree, because you included another revert |
| 13:09.53 | Pali | commit d5b97aa802ff650878c059dda8ddba31433e01f6 |
| 13:09.54 | freemangordon | Pali: could pull my branch now and make a diff |
| 13:10.04 | Pali | freemangordon: already did |
| 13:10.06 | freemangordon | if tose are equal, I won;t re-clone |
| 13:10.07 | freemangordon | ok |
| 13:10.14 | freemangordon | well, I'll kep my tree |
| 13:10.17 | freemangordon | *keep |
| 13:10.25 | Pali | [15:06:25] <Pali> freemangordon: btw, I included all your commits into v3.10-n900 branch |
| 13:10.34 | freemangordon | I prefer to not touch yours, as I do lots of changes usually |
| 13:10.42 | Pali | [15:07:37] <Pali> freemangordon: I think you can now remove your modified branch |
| 13:10.43 | Pali | [15:07:52] <Pali> because it has all and same commits as which are in global one |
| 13:11.03 | Pali | ok, just reset --hard to HEAD of global tree |
| 13:11.13 | freemangordon | and I prefer you to cherrypick when something is needed. |
| 13:11.16 | freemangordon | ok |
| 13:12.37 | freemangordon | "HEAD is now at 69346f6 board-rx51-camera: support both cameras in board code" |
| 13:12.46 | *** join/#maemo-ssu mkaindl (~mkaindl@ama-dablam.markus-kaindl.de) |
| 13:15.38 | Pali | de13db43ace67440a5e8d3bcc128d3bcfe0fd967 |
| 13:15.38 | Pali | this is ^^^ |
| 13:17.54 | *** join/#maemo-ssu DrCode (~DrCode@gateway/tor-sasl/drcode) |
| 13:20.17 | Pali | freemangordon: HEAD of global tree not HEAD of your local repo |
| 13:24.25 | freemangordon | Pali: hmm? |
| 13:24.37 | freemangordon | how to do that and why? |
| 13:25.25 | jonwil | 'figuring out libdevlock is harder than I thought |
| 13:25.41 | Pali | why? because your tree contains only irrelevant commits (commit1, revert commit1, commit2, revert commit2 ...) |
| 13:25.50 | Pali | how? git reset --hard de13db43ace67440a5e8d3bcc128d3bcfe0fd967 |
| 13:26.09 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 13:26.14 | *** join/#maemo-ssu arcean (~arcean@aaeq134.neoplus.adsl.tpnet.pl) |
| 13:26.42 | Pali | jonwil: look at diablo mce |
| 13:26.54 | Pali | diablo mce has code which was moved into libdevlock |
| 13:26.58 | jonwil | ok |
| 13:27.14 | Pali | code for checking security code |
| 13:27.33 | Pali | and cal structure of that security code storage |
| 13:28.25 | jonwil | ok |
| 13:31.34 | jonwil | I shall soon have devlock.h usable :) |
| 13:39.01 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 13:41.25 | *** join/#maemo-ssu discopig (~discopig@unaffiliated/discopig) |
| 13:44.52 | jonwil | hmmm, need to see who else uses libdevlock :) |
| 13:45.11 | *** join/#maemo-ssu arcean_ (~arcean@aacx222.neoplus.adsl.tpnet.pl) |
| 13:50.24 | freemangordon | Pali: Nokia-N900:~# ls -al /dev/hwrng |
| 13:50.25 | freemangordon | crw-rw---- 1 root root 10, 183 Jan 1 1970 /dev/hwrng |
| 13:50.37 | freemangordon | seems to work |
| 13:50.46 | freemangordon | ca filled my reminal with garbage :D |
| 13:50.52 | freemangordon | *terminal |
| 13:52.16 | Pali | ok :-) |
| 13:52.23 | Pali | check also dmesg if there is no error |
| 13:52.37 | freemangordon | what error to look for? |
| 13:52.50 | Pali | when you started cat |
| 13:52.58 | Pali | for omap3 rom rng |
| 13:52.59 | freemangordon | oh, you mean while doing cat? |
| 13:53.01 | freemangordon | ok |
| 13:53.42 | freemangordon | Pali: no errors |
| 13:54.09 | Pali | ok, then it working... |
| 13:56.44 | Pali | freemangordon: can I rebase n900 patches on top of 3.11/3.12-rc now or wait until you fix modem/camera problems? |
| 13:56.51 | freemangordon | rebase |
| 13:56.56 | freemangordon | *NOW* :) |
| 13:57.04 | Pali | ok :-) going to look at it |
| 13:57.31 | Pali | there is 3.12-rc1 |
| 13:57.55 | freemangordon | bravely do git pull --rebase :D |
| 13:58.38 | Pali | haha :D rather I clean some commit, revert commit patches... |
| 13:59.07 | freemangordon | hmm, could you pull master? |
| 13:59.46 | freemangordon | wel, well, I'll wait you to merge it |
| 13:59.57 | freemangordon | scratch that (git pull) |
| 14:00.25 | freemangordon | Pali: BTW ay idea why power button doesn;t wotk? |
| 14:00.33 | freemangordon | *any |
| 14:00.43 | Pali | in mce? |
| 14:01.00 | Pali | because mce has hardcoded some /dev/input* path? |
| 14:01.06 | freemangordon | hmm, ok |
| 14:01.15 | Pali | and devices in input are randomly generated? |
| 14:13.34 | *** join/#maemo-ssu peetah (~peetah@cha92-9-82-236-202-86.fbx.proxad.net) |
| 14:15.53 | freemangordon | Pali: https://gitorious.org/linux-n900/freemangordons-linux-n900/source/a6d3bd274b85218bf7dda925d14db81e1a8268b3:drivers/input/misc/twl4030-pwrbutton.c#L69 |
| 14:16.28 | freemangordon | on 2.6.28 it is triton2-pwrbutton |
| 14:16.36 | Pali | could be |
| 14:21.29 | *** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au) |
| 14:22.46 | DocScrutinizer05 | guys, what's the headline to put on top of the report about your work of last 2 weeks? What's the goal? |
| 14:23.16 | freemangordon | maemo5 working with upstream kernel? |
| 14:23.17 | DocScrutinizer05 | is there a plan? |
| 14:23.30 | freemangordon | yep, make it working :D |
| 14:23.48 | DocScrutinizer05 | hmm, what for? |
| 14:24.02 | freemangordon | we'll see |
| 14:24.06 | DocScrutinizer05 | mhm |
| 14:25.58 | freemangordon | dammit, mce had those names hardcoded :( |
| 14:26.40 | freemangordon | Pali: how to deal with that? |
| 14:27.13 | Pali | somehow change mce code |
| 14:27.19 | freemangordon | I mean - can;t we fix it in the kernel, with some config option? would upstream accept that? |
| 14:27.36 | Pali | no, this is problem in userspace application |
| 14:29.38 | freemangordon | well, yes, but can't we use INPUT_TWL4030_PWRBUTTON_NAME to define a name? |
| 14:29.44 | freemangordon | for example |
| 14:30.18 | freemangordon | the same for vibra, etc |
| 14:30.29 | freemangordon | Pali: ^^^ |
| 14:30.54 | Pali | freemangordon: ask input guys |
| 14:30.59 | Pali | I think asnwer will be no |
| 14:31.29 | freemangordon | Pali: any way to make an alias? |
| 14:31.43 | Pali | do not know any way |
| 14:31.51 | freemangordon | ok, I'll try to find |
| 14:32.11 | DocScrutinizer05 | well, isn't that exactly one of the main reasons why we didn't even think of going to newer kernel versions, for e.g. KP? |
| 14:32.35 | DocScrutinizer05 | API compatibility |
| 14:32.40 | freemangordon | it is |
| 14:32.52 | DocScrutinizer05 | and the notorious tendency of kernel devels to break that API |
| 14:33.07 | Pali | kernel does not breaking API |
| 14:33.12 | Pali | and if yes, report it |
| 14:33.26 | Pali | this is not API, but mce bad sw... |
| 14:33.43 | freemangordon | Pali: I wouldn't blame mce in that case |
| 14:33.54 | DocScrutinizer05 | another even more serious reason: power management customary optimizations done by Nokia to maemo kernel, that never can go upstream |
| 14:33.56 | Pali | ok |
| 14:34.21 | DocScrutinizer05 | Pali: HAHA |
| 14:34.38 | DocScrutinizer05 | changing sysfs paths is not "breaking API"? |
| 14:34.39 | freemangordon | because someone decided that "twl4030_pwrbutton" is more cool than "triton2-powerbuton" |
| 14:35.10 | freemangordon | BTW I don;t think ^^^ is part of the API aiui |
| 14:35.12 | DocScrutinizer05 | ooh, devfs paths then, in this case, eh? |
| 14:35.20 | freemangordon | no, patch are just fine |
| 14:35.49 | freemangordon | it is the "name" property that has changed, and it is up to device drier to set that |
| 14:35.53 | freemangordon | *path |
| 14:36.14 | DocScrutinizer05 | so what? |
| 14:36.21 | DocScrutinizer05 | API change |
| 14:36.35 | freemangordon | no, driver change |
| 14:36.48 | DocScrutinizer05 | doesn't make any difference in my book |
| 14:37.33 | DocScrutinizer05 | kernel devel sophism: this is not a bug, this is a grammatical change |
| 14:38.56 | freemangordon | DocScrutinizer05: whatever it is, we have to deal with it |
| 14:39.02 | DocScrutinizer05 | 1.: declare own prior work "buggy" 2.: declare new API a "bugfix" 3.: declare all apps that depend on the supposed former "bug2 flawed |
| 14:39.51 | DocScrutinizer05 | we have to deal with it for what exactly? |
| 14:40.25 | DocScrutinizer05 | ooh, <freemangordon> we'll see |
| 14:47.13 | DocScrutinizer05 | anyway I will point bitching nagging users to this, like estel and ivgalvez who claim that "cssu team already has way too much on their plate to get it done in any acceptable time frame" |
| 15:16.01 | *** join/#maemo-ssu arcean_ (~arcean@aacx222.neoplus.adsl.tpnet.pl) |
| 15:19.58 | freemangordon | Pali: hmm, can't we LD_PELOAD? |
| 15:20.15 | *** join/#maemo-ssu iDont (~iDont@ip4da305b4.direct-adsl.nl) |
| 15:20.40 | Pali | which symbol to replace? |
| 15:21.13 | freemangordon | mce does EVIOCGNAME, checking the names in /dev/input |
| 15:21.33 | freemangordon | we can capture ioctl and replace the result with wath mce expects |
| 15:21.35 | Pali | ioctl? |
| 15:21.38 | freemangordon | yep |
| 15:21.42 | Pali | you can |
| 15:21.45 | Pali | why not? |
| 15:22.00 | freemangordon | I know :), but does it makes sense to do it |
| 15:22.13 | freemangordon | *make |
| 15:32.40 | Pali | freemangordon: rebasing patches is complete... I will try to compile kernel now |
| 15:33.42 | *** join/#maemo-ssu discopig (~discopig@unaffiliated/discopig) |
| 15:34.41 | *** join/#maemo-ssu bsdmaniak (~bsdmaniak@std93-20-88-120-139-80.fbx.proxad.net) |
| 16:20.13 | DocScrutinizer05 | why not simply bindmount? |
| 16:20.40 | DocScrutinizer05 | you can do per-process bindmounts that override single filenames |
| 16:21.55 | DocScrutinizer05 | check "man mount" |
| 16:22.45 | DocScrutinizer05 | also see http://www.ibm.com/developerworks/linux/library/l-mount-namespaces/index.html |
| 16:27.54 | Pali | DocScrutinizer05: because of ioctl? |
| 16:28.08 | DocScrutinizer05 | err |
| 16:28.30 | DocScrutinizer05 | how's ioctl not affected by mounts? |
| 16:29.03 | Pali | mce using ioctl and checking some buffer if is same as hardcoded something |
| 16:29.34 | Pali | freemangordon: btrfs is not compiling on 3.12-rc1 :D |
| 16:29.48 | Pali | fs/btrfs/ioctl.c:2802:3: error: implicit declaration of function ‘__put_user_unaligned’ [-Werror=implicit-function-declaration] |
| 16:31.00 | Pali | disabling it |
| 16:47.14 | *** join/#maemo-ssu nox- (noident@freebsd/developer/nox) |
| 16:57.55 | *** join/#maemo-ssu sailus (~sailus@valkosipuli.retiisi.org.uk) |
| 16:58.27 | DocScrutinizer05 | I dunno what exactly you're doing, and why, but at least the FPTP paradigm been to keep API as identical as possible, to allow arbitrary userland stuff to run on the new platform |
| 17:01.13 | DocScrutinizer05 | s/FPTP/FPTF/ |
| 17:02.58 | *** join/#maemo-ssu M13 (~MirandaLS@170.133-224-87.telenet.ru) |
| 17:03.56 | Sicelo | ~fptp |
| 17:04.59 | Sicelo | ~fptf |
| 17:05.27 | Sicelo | ~sicelo |
| 17:15.05 | Pali | freemangordon: v3.12-rc1-n900 branch is on gitorious |
| 17:27.44 | freemangordon | Pali: great |
| 17:27.55 | freemangordon | does it boot? |
| 17:28.04 | Pali | in qemu yes |
| 17:28.13 | Pali | not tested on real n900 yet |
| 17:28.28 | Pali | and qemu now also see sound card |
| 17:28.53 | Pali | I do not have free space on rootfs for modules :D |
| 17:29.01 | Pali | need something to clean |
| 17:29.40 | freemangordon | ok, going to test it on the real device |
| 17:32.00 | Pali | ah groff-base package is not optified |
| 17:32.07 | Pali | and man-db-n900 too |
| 17:40.59 | *** join/#maemo-ssu arcean_ (~arcean@aacx222.neoplus.adsl.tpnet.pl) |
| 17:49.11 | freemangordon | Pali: which branch did you base your v.3.12-rc1-n900? |
| 17:49.34 | freemangordon | master or v3.10? |
| 17:50.31 | Pali | v3.10 |
| 17:50.47 | freemangordon | ok |
| 18:33.58 | Pali | freemangordon: did you try new branch? |
| 18:34.16 | freemangordon | still doing pull :) |
| 18:34.23 | freemangordon | i re-cloned the repo |
| 18:34.45 | freemangordon | as git failed to merge your(upstream :P) remote |
| 18:34.54 | freemangordon | git in SB is pretty old |
| 18:35.56 | freemangordon | Pali: hmm: https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/7b2e1277598e4187c9be3e61fd9b0f0423f97986 |
| 18:36.13 | freemangordon | https://gitorious.org/linux-n900/freemangordons-linux-n900/commit/79324a2d5bf6baa25dcb2de58ba5c79410bb365c |
| 18:36.37 | Pali | freemangordon: you cannot merge it |
| 18:36.41 | Pali | it is new branch |
| 18:36.52 | freemangordon | yep, I know |
| 18:37.01 | freemangordon | id did git barnch/git pull |
| 18:37.06 | freemangordon | I did* |
| 18:37.24 | freemangordon | but it failed |
| 18:38.00 | Pali | $ git fetch <name_of_remote> |
| 18:38.09 | freemangordon | did that too |
| 18:38.29 | Pali | $ git checkout -b v3.12-rc1-n900 <name_of_remote>/v3.12-rc1-n900 |
| 18:38.50 | Pali | this should work (if you do not have branch v3.12-rc1-n900 already) |
| 18:38.56 | freemangordon | oh, right :) |
| 18:39.07 | freemangordon | well, will remember that for the next time |
| 18:39.14 | freemangordon | anyway, finished |
| 18:39.21 | freemangordon | going to build it |
| 18:39.45 | Pali | $ git rev-parse HEAD |
| 18:39.48 | Pali | 76a9602a630b816822c540c50e2979f7cdde27cb |
| 18:39.56 | Pali | freemangordon: check this ^^^ |
| 18:41.16 | Pali | freemangordon: can you check if we need CONFIG_ARM_ERRATA_458693 and CONFIG_ARM_ERRATA_460075 ? |
| 18:41.27 | Pali | because both options are since 3.10 disabled |
| 18:41.41 | freemangordon | afaik the only errata we need is thumb one |
| 18:41.46 | freemangordon | but ok, will check |
| 18:42.02 | freemangordon | Pali: did you look at the above commits? |
| 18:42.12 | Pali | yes, what is problem? |
| 18:43.08 | *** join/#maemo-ssu arcean_ (~arcean@aacx222.neoplus.adsl.tpnet.pl) |
| 18:43.45 | freemangordon | oh, scratch that, upstream is for 3630 |
| 18:45.33 | freemangordon | 458693 is for Cortex-A8 (r2p0) we are on r2p3 iirc |
| 18:45.58 | freemangordon | the same for the 460075 |
| 18:46.39 | freemangordon | so, those should be disabled |
| 18:46.52 | freemangordon | Pali: ^^^ |
| 18:47.30 | Pali | those are disabled since 3.10 |
| 18:47.42 | freemangordon | good |
| 18:48.15 | Pali | removing master branch from linux-n900 tree because is useless for us |
| 18:48.26 | freemangordon | :nod: |
| 18:48.49 | freemangordon | though... |
| 18:48.57 | freemangordon | why not pull upstream? |
| 18:49.02 | Pali | I set HEAD of linux-n900 tree to v3.12-rc1-n900 |
| 18:49.46 | Pali | freemangordon: because master is already and periodical mirror is pointless |
| 18:49.54 | freemangordon | ok |
| 18:50.15 | freemangordon | waits for mudules build to finish |
| 18:50.21 | freemangordon | modules even |
| 18:50.26 | Pali | if you or somebody else need to see unreleased (v3.12-rc2 ..rc3...) then he will look at linus tree |
| 18:55.41 | Pali | I pushed to linux-n900 tree also base tags where branches starting |
| 18:55.55 | Pali | you can git fetch <remote> to get these tags |
| 18:56.19 | freemangordon | ok |
| 19:09.04 | DocScrutinizer05 | ~FPTF is the Fremantle Porting Task Force, see http://talk.maemo.org/showthread.php?t=91308 |
| 19:09.04 | infobot | DocScrutinizer05: okay |
| 19:09.25 | freemangordon | Pali: looks like it boots :) |
| 19:10.02 | Pali | freemangordon: check if lp5523 led with MCE working |
| 19:10.41 | DocScrutinizer05 | first check lp5523 working ;-) |
| 19:10.47 | freemangordon | ok |
| 19:10.53 | DocScrutinizer05 | then patch lp5523 |
| 19:10.59 | DocScrutinizer05 | ;-) |
| 19:11.03 | Pali | slide lock and wait if you see white color... |
| 19:11.35 | Pali | DocScrutinizer05: lp5523 was rewritten... |
| 19:11.44 | Pali | and has new common interface for lp55xx chips |
| 19:12.01 | DocScrutinizer05 | still fsckdup |
| 19:12.12 | Pali | and freemangordon is first who will test it :D |
| 19:12.16 | freemangordon | Pali: no :( |
| 19:12.22 | DocScrutinizer05 | of course |
| 19:12.25 | DocScrutinizer05 | see ^^^ |
| 19:12.35 | freemangordon | though, I'll powercycle just to be sure |
| 19:12.47 | Pali | nice, another non working part... I will write that TI devs |
| 19:12.58 | DocScrutinizer05 | MCE relies on NOKIA idiosyncrasies^htics |
| 19:13.05 | Pali | last time they wrote me that it should be fixed in 3.11 |
| 19:13.25 | freemangordon | Pali: wait |
| 19:13.33 | Pali | ok :-) |
| 19:13.37 | freemangordon | let me check why it not working first |
| 19:14.07 | Pali | freemangordon: is keyboard ligh working? |
| 19:14.10 | DocScrutinizer05 | I honestly doubt that led:r, led:g, led:b and kbd:1..kbd:6 have any chance to go upstream |
| 19:14.24 | Pali | DocScrutinizer05: already upstreamed by me :D |
| 19:14.31 | freemangordon | :nod: |
| 19:14.34 | freemangordon | :P |
| 19:14.37 | Pali | if they not deleted it |
| 19:15.16 | freemangordon | hmm, et8ek8 oopses, maybe again some mess on my side |
| 19:15.33 | DocScrutinizer05 | any general purpose upstream LP5523 driver may only have led:1 .. led:9 |
| 19:15.48 | DocScrutinizer05 | and no led:r,g,b and kbd:1..6 |
| 19:15.56 | freemangordon | DocScrutinizer05: names come from the board code |
| 19:16.12 | freemangordon | iirc |
| 19:16.14 | DocScrutinizer05 | then it may work, but I heard board code is obsolete? |
| 19:16.46 | DocScrutinizer05 | wasn't that the DT hyoe now? |
| 19:16.52 | DocScrutinizer05 | hype even |
| 19:19.38 | Pali | DocScrutinizer05: led:r ... kbd:1 .. names are in board data file |
| 19:19.54 | DocScrutinizer05 | ok then |
| 19:20.03 | DocScrutinizer05 | check if the driver works |
| 19:20.27 | DocScrutinizer05 | echo 188 >./LED:R/brightness |
| 19:20.28 | Pali | when board data file will be obsolated by DT, then also these names will be moved to DT |
| 19:20.43 | Pali | names are static data too, so it is ideal for DT |
| 19:20.58 | DocScrutinizer05 | yep |
| 19:21.37 | Pali | freemangordon: maybe there was rebase error in et8ek8? |
| 19:22.09 | DocScrutinizer05 | freemangordon: http://privatepaste.com/bb9fe98506 |
| 19:22.30 | freemangordon | Pali: http://pastebin.com/eftfLEHs |
| 19:22.56 | *** join/#maemo-ssu _nicolai_ (~nicolai@pop8-126.catv.wtnet.de) |
| 19:23.24 | DocScrutinizer05 | http://privatepaste.com/55a214695f |
| 19:24.17 | Pali | what? /sys/class/leds/twl4030:vibrator/brightness |
| 19:24.28 | freemangordon | DocScrutinizer05: yep, have those, besides /sys/class/backlight/acx565akm/brightness and /sys/class/leds/twl4030:vibrator/brightness |
| 19:24.29 | freemangordon | Pali: no :) |
| 19:24.40 | DocScrutinizer05 | echo 111 > /sys/class/leds/lp5523\:g/brightness |
| 19:25.20 | freemangordon | blinks green for a split second |
| 19:25.43 | DocScrutinizer05 | IroN900:~# echo 2 >/sys/class/leds/lp5523\:g/led_current |
| 19:25.57 | DocScrutinizer05 | freemangordon: then MCE already resets it |
| 19:26.00 | DocScrutinizer05 | :-D |
| 19:26.26 | freemangordon | constant green |
| 19:26.28 | DocScrutinizer05 | stop mce; echo 111 > /sys/class/leds/lp5523\:g/brightness |
| 19:26.36 | freemangordon | with mce running |
| 19:26.49 | DocScrutinizer05 | well, until mce gets *any* event |
| 19:26.50 | freemangordon | but still no notifications |
| 19:27.13 | freemangordon | no, mce doesn't seem to control it |
| 19:27.19 | DocScrutinizer05 | which notifications? those in /etc/mce/mce.ini? ;-) |
| 19:27.25 | freemangordon | yep |
| 19:29.00 | DocScrutinizer05 | when green for a split second, then *somebody* seems to reset it |
| 19:29.04 | freemangordon | Pali: or there are some changes in smiareglist |
| 19:29.10 | DocScrutinizer05 | I dunno of anything but MCE doing that |
| 19:29.20 | freemangordon | DocScrutinizer05: it was before I did echo 2 .. |
| 19:29.34 | DocScrutinizer05 | lol |
| 19:31.01 | DocScrutinizer05 | http://privatepaste.com/ed28f5e2d4 |
| 19:32.10 | freemangordon | http://pastebin.com/t49BeY6t |
| 19:32.13 | DocScrutinizer05 | upstream driver might fail to init those |
| 19:32.43 | DocScrutinizer05 | 12 and 50, WOW |
| 19:32.45 | *** join/#maemo-ssu Vlad_on_the_road (~Vlad_on_t@ip-66.net-82-216-1.versailles2.rev.numericable.fr) |
| 19:32.46 | DocScrutinizer05 | blinding |
| 19:32.58 | DocScrutinizer05 | and possibly dangerous for the LED |
| 19:33.17 | DocScrutinizer05 | I once evaluated that, see wiki |
| 19:33.34 | freemangordon | oh, kbd leds shine when I open the kbd |
| 19:33.42 | freemangordon | I guess this means the driver works :) |
| 19:34.02 | DocScrutinizer05 | I hope they *stop* shining when you close the kbd ;-) |
| 19:34.14 | freemangordon | they do |
| 19:34.19 | DocScrutinizer05 | \o/ |
| 19:34.27 | freemangordon | well, I am in RD mode, so part of the continues :) |
| 19:34.29 | DocScrutinizer05 | toldya mce works |
| 19:34.43 | freemangordon | but why indicator led does not? |
| 19:34.51 | DocScrutinizer05 | I tend to blame your mce.ini |
| 19:34.57 | freemangordon | it is stock |
| 19:35.01 | DocScrutinizer05 | hmm |
| 19:35.05 | DocScrutinizer05 | nfc then |
| 19:35.25 | freemangordon | yep, this is my devel device I try to keep as close to stock config as possible |
| 19:36.36 | DocScrutinizer05 | check if bme selects according pattern via dbus |
| 19:36.58 | DocScrutinizer05 | try charging |
| 19:37.16 | DocScrutinizer05 | change settings for indicator LED |
| 19:37.42 | DocScrutinizer05 | select a particular high prio pattern via dbus-send |
| 19:37.50 | freemangordon | hmm, seems mce doesn't know the screen is locked |
| 19:37.51 | DocScrutinizer05 | ~phone-control |
| 19:37.57 | DocScrutinizer05 | ~phonecontrol |
| 19:37.57 | infobot | methinks phonecontrol is http://wiki.maemo.org/Phone_control |
| 19:38.11 | DocScrutinizer05 | exactly what I meant |
| 19:39.00 | DocScrutinizer05 | indicator pattern is engine state defendent |
| 19:39.04 | DocScrutinizer05 | dependent |
| 19:39.48 | DocScrutinizer05 | mce has quite a nasty set of functions to change engine state |
| 19:40.01 | DocScrutinizer05 | on God knows which events |
| 19:40.15 | freemangordon | :) |
| 19:41.01 | freemangordon | DocScrutinizer05: what was that selftest thingie? |
| 19:41.12 | DocScrutinizer05 | press power button shortly and see if white LED ramps up |
| 19:41.41 | DocScrutinizer05 | freemangordon: selftest gets executed during module load and enables all LED and checks current flow |
| 19:41.58 | DocScrutinizer05 | for a good/nogood test |
| 19:42.21 | DocScrutinizer05 | it actually does a selftest of all LED |
| 19:42.22 | freemangordon | Nokia-N900:~# cat /sys/devices/platform/omap_i2c.2/i2c-2/2-0032/selftest |
| 19:42.22 | freemangordon | OK |
| 19:42.37 | *** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl) |
| 19:42.54 | DocScrutinizer05 | you even should see LEDs flashing on cat /sys/devices/platform/omap_i2c.2/i2c-2/2-0032/selftest |
| 19:43.01 | freemangordon | I see that |
| 19:43.08 | *** join/#maemo-ssu _rd (~rd@p5088FD88.dip0.t-ipconnect.de) |
| 19:43.17 | freemangordon | well, seems mce is just unaware thet the screen is locked |
| 19:43.23 | freemangordon | *that |
| 19:43.29 | freemangordon | going to check what it wants |
| 19:43.53 | DocScrutinizer05 | freemangordon: sorry, path not valid on my platform |
| 19:44.07 | freemangordon | hmm, it should be |
| 19:44.13 | DocScrutinizer05 | kernel damagers' API change |
| 19:44.26 | freemangordon | DocScrutinizer05: the path is the same on 2.6.28 |
| 19:44.48 | DocScrutinizer05 | IroN900:~# find /sys -name selftest |
| 19:44.50 | DocScrutinizer05 | /sys/class/i2c-adapter/i2c-2/2-0032/selftest |
| 19:44.56 | DocScrutinizer05 | no it isn't |
| 19:44.59 | DocScrutinizer05 | evidently |
| 19:45.07 | freemangordon | do you have /sys/class/i2c-adapter/i2c-2/2-0032 ? |
| 19:45.13 | DocScrutinizer05 | no |
| 19:45.29 | DocScrutinizer05 | err yep |
| 19:45.33 | freemangordon | :) |
| 19:45.41 | freemangordon | maybe no selftest there |
| 19:45.50 | DocScrutinizer05 | yep |
| 19:45.54 | DocScrutinizer05 | selftest there |
| 19:45.58 | freemangordon | oh |
| 19:46.06 | DocScrutinizer05 | IroN900:~# ll /sys/class/i2c-adapter/i2c-2/2-0032/selftest |
| 19:46.08 | DocScrutinizer05 | -r--r--r-- 1 root root 4096 2013-09-22 21:45 /sys/class/i2c-adapter/i2c-2/2-0032/selftest |
| 19:46.14 | freemangordon | so, the same |
| 19:46.23 | dos1 | <PROTECTED> |
| 19:46.32 | DocScrutinizer05 | but I DO NOT have /sys/devices/platform/omap_i2c.2 |
| 19:46.53 | DocScrutinizer05 | IroN900:~# ll /sys/devices/platform/omap_i2c.2 |
| 19:46.54 | DocScrutinizer05 | ls: Zugriff auf /sys/devices/platform/omap_i2c.2 nicht möglich: No such file or directory |
| 19:47.13 | freemangordon | weird |
| 19:47.35 | DocScrutinizer05 | kernel damagers' API change |
| 19:47.45 | freemangordon | oh, it is a symlink here |
| 19:48.23 | DocScrutinizer05 | LOL |
| 19:48.41 | dos1 | I think /sys/class is always a prefered "API" to be used by userspace |
| 19:48.53 | freemangordon | Nokia-N900:~# ls -al /sys/class/i2c-adapter/i2c-2/2-0032/selftest |
| 19:48.54 | freemangordon | -r--r--r-- 1 root root 4096 Jan 1 06:21 /sys/class/i2c-adapter/i2c-2/2-0032/selftest |
| 19:49.10 | freemangordon | so, the same |
| 19:49.25 | freemangordon | cat that |
| 19:49.31 | dos1 | and anything in /sys/class is always duplicated somewhere |
| 19:49.33 | freemangordon | :nod: |
| 19:49.42 | dos1 | and that second place can change, but /sys/class is rather still |
| 19:49.51 | DocScrutinizer05 | anyway `cat /sys/class/i2c-adapter/i2c-2/2-0032/selftest` nicely flashes all LED here |
| 19:50.02 | freemangordon | same here |
| 19:50.06 | DocScrutinizer05 | even nicely animated |
| 19:50.17 | freemangordon | the problem seems to be with kb_lock gpio |
| 19:50.31 | DocScrutinizer05 | hmhm |
| 19:50.38 | DocScrutinizer05 | check dmesg |
| 19:50.56 | dos1 | there should be sysfs node to check kb_lock state |
| 19:51.00 | freemangordon | no, it is fine as the screen gets blanked |
| 19:51.02 | DocScrutinizer05 | [124629.358856] slide (GPIO 71) is now closed |
| 19:51.04 | DocScrutinizer05 | [125162.695770] slide (GPIO 71) is now open |
| 19:51.05 | Pali | freemangordon: can you try LED Patterns settings applet? |
| 19:51.28 | Pali | there is test button :-) |
| 19:51.30 | freemangordon | Jan 1 06:32:42 Nokia-N900 kernel: [ 2246.508117] kb_lock (GPIO 113) is now closed |
| 19:51.31 | freemangordon | Jan 1 06:32:42 Nokia-N900 kernel: [ 2246.751708] kb_lock (GPIO 113) is now open |
| 19:51.36 | freemangordon | whaaaat? |
| 19:51.39 | dos1 | freemangordon: strace mce then? |
| 19:51.41 | DocScrutinizer05 | err |
| 19:51.47 | DocScrutinizer05 | kbd_lock? |
| 19:51.50 | freemangordon | Pali: driver seems to work |
| 19:51.55 | freemangordon | DocScrutinizer05: yep |
| 19:52.00 | DocScrutinizer05 | I meant *slide* |
| 19:52.04 | DocScrutinizer05 | well |
| 19:52.09 | Pali | freemangordon: but API for led patterns was broken in 3.10 |
| 19:52.10 | dos1 | freemangordon: it's closed *while* pressing, and open when you release it |
| 19:52.16 | freemangordon | this is what I get when I lock with the slide :D |
| 19:52.19 | DocScrutinizer05 | [125695.666809] kb_lock (GPIO 113) is now closed |
| 19:52.21 | DocScrutinizer05 | [125696.510192] kb_lock (GPIO 113) is now open |
| 19:52.22 | DocScrutinizer05 | for you |
| 19:52.31 | Pali | and TI devs wrote me that they tried to fix it in 3.11 |
| 19:53.02 | freemangordon | DocScrutinizer05: what is slide and what is kblock? |
| 19:53.11 | DocScrutinizer05 | slide is kbd |
| 19:53.11 | Pali | http://maemo.org/packages/package_instance/view/fremantle_extras-devel_free_armel/led-pattern-editor/0.0.5-1/ |
| 19:53.20 | freemangordon | Pali: ok |
| 19:53.21 | DocScrutinizer05 | kb_lock is lock switch |
| 19:53.26 | freemangordon | DocScrutinizer05: this is fine (the slide) |
| 19:53.54 | Pali | freemangordon: one of GPIO switch cannot be registered to nokia special omap-gpio-switch driver |
| 19:53.59 | freemangordon | what I meant is that when device gets locked with the lock switch, it seems mce don;t understabd that |
| 19:54.04 | Pali | because is used by another driver |
| 19:54.10 | freemangordon | Pali: which one? |
| 19:54.24 | Pali | and I patched that omap gpio driver to not fail loading if some gpio is in use |
| 19:54.28 | DocScrutinizer05 | tklock.so? |
| 19:54.32 | Pali | freemangordon: I do not remember, but maybe slide |
| 19:54.33 | freemangordon | hmm, it is in dmesg |
| 19:54.38 | freemangordon | is going to check |
| 19:54.51 | Pali | look to dmesg, it should report if it failed |
| 19:54.59 | freemangordon | Pali: could you check why et8ek8 oopses? |
| 19:55.15 | Pali | freemangordon: already looked but I have no idea... |
| 19:55.23 | DocScrutinizer05 | freemangordon: wait for mce to lock on timeout |
| 19:55.32 | freemangordon | hmm, will do depmad -a and will retry |
| 19:55.36 | freemangordon | *depmod |
| 19:55.38 | Pali | if everything worked with 3.10 and now not working with 3.12 there will be some problem somewhere... |
| 19:55.57 | Pali | I used same et8ek8 patches for 3.10 and 3.12 |
| 19:56.45 | DocScrutinizer05 | fetches a rope, TV on, newsflash, then shoots himself with that roppe |
| 19:57.58 | DocScrutinizer05 | only mitigating news: FDP is out |
| 19:58.04 | DocScrutinizer05 | **finally** |
| 20:01.36 | freemangordon | Pali: hmm, don;t see any failed gpio, could you please gimme some more info? |
| 20:02.17 | Pali | freemangordon: look for string "gpio-switch" |
| 20:02.36 | freemangordon | ok |
| 20:02.40 | freemangordon | found it |
| 20:02.57 | Pali | gpio-switch: gpio_reguest failed for |
| 20:03.04 | freemangordon | yep |
| 20:03.18 | freemangordon | this is not our gpio |
| 20:03.34 | Pali | which gpio failed? |
| 20:03.40 | freemangordon | 177 |
| 20:03.49 | freemangordon | gpio-switch: gpio_reguest failed for headphone 177 |
| 20:04.00 | Pali | #define RX51_GPIO_HEADPHONE177 |
| 20:04.15 | Pali | ok, it looks like that headphone gpio is controlled by some input driver |
| 20:04.31 | freemangordon | :nod: |
| 20:04.41 | freemangordon | I guess we should remove it from the board data |
| 20:04.42 | Pali | by kernel input system |
| 20:05.00 | Pali | freemangordon: not because some maemo daemon/app failing |
| 20:05.13 | Pali | this is reason why added read only data from sysfs |
| 20:05.29 | freemangordon | ok |
| 20:05.39 | Pali | also gpio-switch is not in upstream and never will be |
| 20:05.46 | freemangordon | ok :) |
| 20:08.00 | DocScrutinizer05 | umm |
| 20:08.26 | DocScrutinizer05 | headphone switch needed for ped/PA/alsa |
| 20:11.58 | Pali | for prolog is is needed |
| 20:12.06 | Pali | or maybe MCE? |
| 20:13.18 | *** join/#maemo-ssu sunny_s (~sunny_s@business-092-079-020-027.static.arcor-ip.net) |
| 20:15.22 | *** join/#maemo-ssu _rd (~rd@p5088FD88.dip0.t-ipconnect.de) |
| 20:22.28 | DocScrutinizer05 | tbh I dunno |
| 20:24.05 | DocScrutinizer05 | might be sth like KERNEL(177)->kevent->HALD_ADDON_donnowhat|KE_RECV->HAL->dbus->PED |
| 20:24.29 | DocScrutinizer05 | or even ->MCE->PED |
| 20:24.37 | DocScrutinizer05 | who knows |
| 20:33.42 | freemangordon | hmm: |
| 20:33.44 | freemangordon | Nokia-N900:~# cat /sys/devices/platform/omap_i2c.2/i2c-2/2-0032/leds/lp5523\:b/led_current |
| 20:33.44 | freemangordon | 47 |
| 20:34.04 | freemangordon | this happens when there is a notification |
| 20:34.11 | freemangordon | DocScrutinizer05: ^^^ |
| 20:41.04 | dos1 | reminds me of AUX LED in GTA02 taking 50mA instead of 5 in early revisions |
| 20:41.17 | dos1 | but that was in hw, so just small ot ;) |
| 20:44.20 | *** join/#maemo-ssu sailus (~sailus@valkosipuli.retiisi.org.uk) |
| 20:46.04 | freemangordon | DocScrutinizer05: seems mce calcualtes brightness on the opposite to what the driver expects |
| 20:46.30 | freemangordon | or something like that |
| 20:53.12 | freemangordon | hmm, led-pattern-editor doesn't work either |
| 21:06.20 | *** join/#maemo-ssu Martix (~martix@4.177.broadband3.iol.cz) |
| 21:08.01 | *** join/#maemo-ssu Martix_ (~martix@4.177.broadband3.iol.cz) |
| 21:30.43 | DocScrutinizer05 | freemangordon: there's a clear threshold of ALS above which indLED is dimmed down (by either current_led or brightnes [=PWM]) to near invisible brightness |
| 21:31.41 | DocScrutinizer05 | doesn't exactly look like "calculated" to me - rather I think some fool decided to dim *all* LED when actually only kbd LEDs were meant to get shut down completely |
| 21:33.07 | DocScrutinizer05 | but street gossip has it that some product manager chairfart gashead specified dimming of indLED above a certain brightness as a product requirement, and so devels implemented it in the least-effort way they could think of |
| 21:34.28 | DocScrutinizer05 | btw NB that led_current is NOT in units of mA |
| 21:35.03 | DocScrutinizer05 | *sigh*, again looking up my own statements in wiki |
| 21:36.51 | DocScrutinizer05 | http://wiki.maemo.org/N900_Hardware_LED |
| 21:37.26 | DocScrutinizer05 | >> |
| 21:37.27 | DocScrutinizer05 | Despite the line |
| 21:37.29 | DocScrutinizer05 | #define LP5523_DEFAULT_CURRENT 50 /* microAmps */ |
| 21:37.30 | DocScrutinizer05 | suggesting different, the units in there are 0.1 mA (100 µA per step is correct). So the driver init default "50" for all LEDs is 5 mA, and mce seems to set current to 0.2 mA (value 2 in the led_current sysnode) for B and G, and either to 0.2 mA or 0.8 mA for R. |
| 21:37.32 | DocScrutinizer05 | << |
| 21:38.38 | DocScrutinizer05 | the comment in ( ) is by Nokia_Employee responsible for LP5523 |
| 21:39.41 | DocScrutinizer05 | >>Probably settings as high as 100 for 10 mA are safe for the RGB LED, higher values might destroy the LED. (yes, correct, the RGB could be damaged from heat if set to more than 10 mA on all 3 LEDs) |
| 21:40.32 | DocScrutinizer05 | quite funny that by guesswork been correct to the mA even for the max current allowed :-) |
| 21:40.46 | DocScrutinizer05 | s/ by/ my/ |
| 21:41.11 | *** join/#maemo-ssu Martix_ (~martix@4.177.broadband3.iol.cz) |
| 21:42.15 | DocScrutinizer05 | freemangordon: actually mce is insane to mess with led_current at all |
| 21:43.06 | DocScrutinizer05 | led_current should get (probably hard-)set to the max allowavle current for the LED, and all the rest be done by brightness sysnode |
| 21:43.54 | DocScrutinizer05 | messing with led_current is strongly deprecated and not really justified by any eationale I'd know of, but meh, that's what we got, obviously |
| 21:44.31 | *** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua) |
| 21:45.34 | *** join/#maemo-ssu Martix_ (~martix@4.177.broadband3.iol.cz) |
| 21:46.56 | DocScrutinizer05 | actually you should set led_current of all 3 LEDs of the 3color indLED so that a) each single LED doesn't see overcurrent b) all three LEDs together shouldn't see exceeding the max continuous *power* consumption allowed for the thermal design, and c) in a ratio to each other so that the indLED shines white on brightness[r,g,b]==255 |
| 21:48.37 | DocScrutinizer05 | otherwise the lp5523 logarithmic mode for brightness will cause hue deviations for "white" < 255 |
| 21:49.20 | *** join/#maemo-ssu Martix_ (~martix@4.177.broadband3.iol.cz) |