| Wizzup | uvos: do you need me to fix the dtb situation, or? | 02:48 |
|---|---|---|
| freemangordon | Wizzup: do we have any plans to upgrade to current stable? | 07:25 |
| freemangordon | bookworm that is | 07:25 |
| freemangordon | Wizzup: so, you say there is "unknown album" in OMP, but there are no songs in it? | 07:36 |
| freemangordon | that could be a tracker issue | 07:36 |
| tmlind | uvos: weird that cpufreq has disappeared | 07:44 |
| tmlind | what, we maybe get mainline mesa for pvr? | 07:45 |
| freemangordon | tmlind: looks like | 07:56 |
| tmlind | nice :) | 07:57 |
| uvos | tmlind: is it the same on your device? | 09:40 |
| tmlind | uvos: what is missing for you? i can do cat /sys/devices/system/cpu/cpu*/cpufreq/scaling_cur_freq just fine looks like | 09:43 |
| uvos | yeah thats missing | 09:43 |
| uvos | ok something got clobbered in the merge then | 09:44 |
| tmlind | sure you have up to date dtb installed? | 09:44 |
| uvos | yeah | 09:44 |
| tmlind | and the cpufreq whatever dt .ko is loaded? | 09:44 |
| uvos | yes its loaded | 09:44 |
| tmlind | weird | 09:45 |
| uvos | it also dosent seam to affect pm, but then again i gues cpufreq only helps when there is partial load | 09:45 |
| uvos | not at idle | 09:45 |
| tmlind | yeah if voltage gets scaled.. | 09:46 |
| ac_laptop | Hello people | 10:12 |
| ac_laptop | there is *definitely* something wrong with my leste install concerning the battery | 10:12 |
| ac_laptop | I've let the device charge all night, when I woke up it was off, I boot it on leste, it boots and says the battery is empty | 10:13 |
| ac_laptop | I replace the battery with an other battery **that I had fully charged the day before**, I boot on leste, it says the battery is empty. | 10:14 |
| ac_laptop | I reboot on fremantle, it shows a full battery. | 10:14 |
| ac_laptop | I suspect maemo sometimes shows me an empty battery when it's actually full (or at least not empty), and shutw down as a security measure | 10:15 |
| ac_laptop | (maemo leste) | 10:15 |
| ac_laptop | I don't know if the battery is not properly calibrated or if something happens when the device is plugged that messes up battery calibration | 10:16 |
| arno11 | ac_laptop: showing an empty battery when it's full is normal if the battery is not calibrated at all. You need 5-10 cycles to get accurate battery data | 10:21 |
| arno11 | *full cycles | 10:21 |
| ac_laptop | arno11: I'm going to do this then. I guess I need to start leste while the device is plugged, wait for a full charge, and then stress it until the battery is empty, and repeat ? | 10:24 |
| ac_laptop | what's the fastest way to empty the battery then ? | 10:24 |
| arno11 | i recommend you to not plug the battery before booting Leste | 10:24 |
| arno11 | it's buggy on n900 | 10:25 |
| ac_laptop | arno11: the thins is, leste believes the battery is empty, right after X is started, it doesn't eve have the time to show all the applets on the desktop, and shots down | 10:25 |
| ac_laptop | thing* | 10:25 |
| ac_laptop | shuts* | 10:26 |
| arno11 | ok | 10:26 |
| arno11 | weird | 10:26 |
| ac_laptop | arno11: I guess battery management shutw down the device when it detects the battery is empty ? as a security measure ? | 10:27 |
| arno11 | yes iirc but don't really know how it works if the battery is not calibrated | 10:28 |
| arno11 | maybe better to start over with a fresh install ? | 10:29 |
| ac_laptop | arno11: it is a fresh install | 10:29 |
| arno11 | ok lol | 10:29 |
| arno11 | i really don't know what's going on. i've never seen this kind of issue (i'm using 2 original batteries) | 10:32 |
| arno11 | for me it is battery related but maybe someone else has another idea | 10:33 |
| arno11 | or maybe something else is shuting down the device | 10:34 |
| arno11 | *shutting | 10:35 |
| arno11 | is swap activated on your device ? | 10:35 |
| arno11 | file system expanded ? | 10:37 |
| arno11 | Was wifi off before the last clean shutdown? | 10:38 |
| arno11 | And which image did you use ? | 10:39 |
| arno11 | or is it a completely new fresh install and first boot ? | 10:42 |
| arno11 | (i ask these questions because maybe we are OOM using last images if swap is still not activated on boot) | 10:43 |
| ac_laptop | arno11: swap is activated, fs is custom-expanded (I grew the partition to 8G), wifi was on the whole time and connected to my network, image is maemo-leste-1.0-armhf-n900-20231112.img | 10:44 |
| ac_laptop | arno11: and it's not first boot, the install was yesterday | 10:45 |
| arno11 | ok | 10:45 |
| ac_laptop | arno11:the OOM at first boot is something I've experienced, but it has nothing to do with the battery | 10:46 |
| arno11 | that's not what i'm saying, just symptoms could be similar | 10:46 |
| uvos__ | we are oom on boot on n900 what!? | 10:52 |
| uvos__ | the only thing we added recently is mafw stuff i think | 10:52 |
| ac_laptop | uvos__: happened to me with the first install on my U1 card. It didn't happen with the U3 card though | 10:56 |
| uvos__ | hmm the sdcard is not related to bein oom | 10:56 |
| Wizzup | freemangordon: ugh @ bookworm, I am not sure | 10:56 |
| uvos__ | or do you mean you where out of disk space? | 10:56 |
| uvos__ | being out of disk space on boot is a solvable problem | 10:57 |
| uvos__ | we can just bump the image size a bit | 10:57 |
| Wizzup | freemangordon: it's a multi month project and I wouldn't like to embark on it now. We should see if itremoves any sw that we depend on, like gtk2 or gconf | 10:57 |
| Wizzup | They probably remove python2 all together | 10:57 |
| arno11 | Wizzup: need your help to check if emmc swap is really activated on boot | 10:57 |
| Wizzup | wait, still catching up on backlog :) | 10:58 |
| Wizzup | best share contents of /etc/fstab first | 10:58 |
| ac_laptop | uvos__: I don't know exactly what happened, it shut down immediately after first boot, either it was because of the lack of swap, or because of the lack of disk space, I suppose | 10:58 |
| uvos__ | it would not shutdown on ether of those | 10:58 |
| uvos__ | being out of disk space can cause wierd issues, but not a shutdown | 10:58 |
| Wizzup | disk full can cause a reboot in some cases | 10:59 |
| uvos__ | bein out of ram would have caused the ire of the oom killer | 10:59 |
| uvos__ | Wizzup: really? how | 10:59 |
| uvos__ | dont we run with dsmes reboots disabled? | 10:59 |
| ac_laptop | uvos__: an Xorg crash causes a shutdown, and I suppose a lack of disk space could cause an Xorg crash | 10:59 |
| uvos__ | xorg crash should not cause a shutdown | 10:59 |
| uvos__ | we disable lifeguard iirc | 10:59 |
| ac_laptop | uvos__: what should it cause ? there's no tty | 11:00 |
| uvos__ | ofc there is tty | 11:00 |
| ac_laptop | I can't get access to it, even when killing Xorg | 11:01 |
| uvos__ | chvt 2 | 11:07 |
| Wizzup | uvos__: oh, that is right @ dsme reboot disabled | 11:15 |
| uvos__ | yeah so a reboot would be very wierd | 11:15 |
| Wizzup | freemangordon: to bookworm to me seems a while off, I would say let's get in beta first and then worry about it | 11:15 |
| Wizzup | because no python2 means no gtk2 for python at all | 11:16 |
| Wizzup | which means we need gtk3 or write our own python gtk2 bindings (afaict) | 11:16 |
| uvos__ | do we have anything that uses gtk2 via python even? | 11:16 |
| Wizzup | yes | 11:16 |
| uvos__ | or are you concerned about porting comunity pacakges? | 11:16 |
| uvos__ | what package? | 11:16 |
| Wizzup | some stuff of hildon I think | 11:16 |
| Wizzup | but also community stuffyes | 11:17 |
| Wizzup | brb | 11:17 |
| ac_laptop | uvos__: had to do it as root, but it works, thanks ! | 11:24 |
| uvos__ | you can also do this via ssh or plug in a keyboard and do ctrl-alt-f2 | 11:29 |
| uvos__ | er nvm n900 dosent have otg so you cant do the latter on n900 | 11:30 |
| arno11 | ac_laptop: i'm currently burning a new img on another usdcard to see what's going on with fresh install | 11:51 |
| uvos__ | arno11: once you flashed, before booting, on pc, see if /etc/no_lg_reboots is present | 11:54 |
| uvos__ | should be a empty file | 11:54 |
| arno11 | argh already booting...any i'll start over to check later | 11:55 |
| arno11 | *anyway | 11:55 |
| uvos__ | check after boot | 11:56 |
| arno11 | ok | 11:56 |
| arno11 | (For the moment, everything is fine) | 11:57 |
| arno11 | battery ok | 11:57 |
| arno11 | ah something is buggy with the display | 11:59 |
| arno11 | wow, first boot is always difficult but now really difficult to use after the first boot | 12:02 |
| uvos__ | details pls | 12:03 |
| arno11 | uvos: no_lg_reboots present and emty | 12:05 |
| arno11 | now it's ok the device is usable, i forgot we have to wait at least 3-4 min to be able to use the desktop | 12:06 |
| arno11 | ah swap is NOT activated | 12:06 |
| uvos__ | arno11: ok | 12:07 |
| arno11 | Wizzup: (for emmc swap) weird | 12:07 |
| arno11 | it explains extreme slowness on first boot and probably ac_laptop OOM | 12:09 |
| Wizzup | arno11: what is the contents of /etc/fstab | 12:11 |
| Wizzup | arno11: can you run dpkg-reconfigure leste-config-n900 after looking, and look again | 12:11 |
| arno11 | argh i already activated emmc swap otherwise it was difficult to use the device | 12:16 |
| Wizzup | arno11: nevermind | 12:16 |
| Wizzup | oh, you mean you added it yourself to /etc/fstab ? | 12:16 |
| arno11 | yes | 12:17 |
| Wizzup | checking if it line exists in the image can be done without dding it again using loopback mount, but the real interesting thing is to check if it gets added (again) with dpkg-reconfigure leste-config-n900 | 12:17 |
| Wizzup | I can't look into this right now, but clearly something is wrong | 12:17 |
| arno11 | yep just wait a min | 12:18 |
| uvos__ | but us ooming with out swap isent good either | 12:19 |
| arno11 | 5 min in fact :) | 12:19 |
| uvos__ | maybe we should have a light hildon meta | 12:19 |
| uvos__ | like without mafw maybe, and maybe some other stuff too | 12:19 |
| uvos__ | i gues outgrowing the n900 is pretty inevitable | 12:20 |
| uvos__ | but im suprised its so big allready | 12:20 |
| arno11 | (to be honest, for a new user, first boot is a nightmare) | 12:24 |
| uvos__ | on n900 anyways :) | 12:25 |
| uvos__ | but its good to see n900 is getting some more testing | 12:25 |
| arno11 | oh yes definitely | 12:26 |
| uvos__ | it deffintaly allways suffers a bit from the d4 being the defacto referance platform | 12:26 |
| arno11 | but to be positive, once configured correctly, everything is ok on n900 | 12:27 |
| arno11 | definitely need to tweak first boot | 12:27 |
| arno11 | Wizzup: dpkg-reconfigure works | 12:28 |
| Wizzup | hm | 12:33 |
| arno11 | and swap is visible | 12:33 |
| Wizzup | uvos__: fremantle will oom without swap too | 12:34 |
| Wizzup | 256MB just isn't enough | 12:34 |
| uvos__ | yeah but its some startup spike | 12:34 |
| uvos__ | on leste atm | 12:34 |
| uvos__ | my d4 uses <200mb ram with everything and some more stuff in installed running | 12:35 |
| arno11 | power button is not working ?!? | 12:35 |
| uvos__ | hmm like not at all? | 12:36 |
| arno11 | yes | 12:36 |
| uvos__ | evtest /dev/input/event* | 12:36 |
| uvos__ | dose it register on the event device its on? | 12:36 |
| arno11 | sorry already run shutdown command | 12:37 |
| ac_laptop | honestly I didn't know about the d4 before hearing about leste. Is the d4 good ? both as a device in general and in terms of leste support ? Alos is there a page listing the nonfree blobs needed on each platform ? | 12:46 |
| uvos__ | ac_laptop: the nonfree blobs are exactly the same: pvrsgx | 12:47 |
| uvos__ | ac_laptop: d4 is slow by modern standarts | 12:47 |
| uvos__ | but its mutch mutch faster than n900, by 5x on the cpu side and it has 4x the ram | 12:47 |
| arno11 | ok so on first reboot everything seems ok and usable | 12:47 |
| uvos__ | its leste support is by far the best of all supported devices | 12:48 |
| uvos__ | the only thing that dosent work is the camera, and that includes good power managment | 12:49 |
| arno11 | bbl | 12:50 |
| ac_laptop | uvos__: the hardware specs are tempting indeed. Is the keyboard pleasant to use ? In terms of freedom I guess it's all the same, unsigned kernel, unsigned bootloader, modem separation ? | 12:52 |
| uvos__ | if find the keyboard vastly superior to the n900 | 12:53 |
| uvos__ | its helped by the device being bigger | 12:53 |
| uvos__ | the bootloader is singed, so is the n900s (nolo) iirc | 12:54 |
| uvos__ | modem seperation is execlent | 12:54 |
| uvos__ | its connected via usb and is powerless to do anything without the cpus help | 12:54 |
| uvos__ | the booloader is however locked | 12:55 |
| uvos__ | but this dosent matter to us as we found a very conveniant way to defeat | 12:55 |
| uvos__ | it | 12:55 |
| ac_laptop | uvos__:ok i thought the bootloader was unsigned on both devices. how are you able to change it then ? | 12:55 |
| uvos__ | on d4 the primary bootloader (mbm) boots a secondary bootloader thats unsigned | 12:55 |
| uvos__ | on n900 its mostly the same the primary bootloader (nolo) boots a secondary bootloader (uboot) | 12:56 |
| uvos__ | but the d4 has more fuckery going on | 12:56 |
| uvos__ | becuase the primay bootloader checks the signature of what is booted | 12:56 |
| uvos__ | but we defeat this | 12:56 |
| uvos__ | to boot our secondary bootloader | 12:56 |
| uvos__ | so it ends up the same in practice | 12:57 |
| ac_laptop | oh but it's a capacitive screen I guess | 12:59 |
| ac_laptop | since it's an android | 12:59 |
| uvos__ | yes | 12:59 |
| sicelo | ac_laptop: you charge the battery outside the device? | 12:59 |
| uvos__ | imo its mutch better than the n900s ts, with not having the option of a stylus being the only downside | 13:00 |
| ac_laptop | sicelo: no, I charge the battery with the device, although both of the batteries have been charged yesterday with the device booted on fremantle, not leste | 13:00 |
| uvos__ | but for thumb useage is vastly superior | 13:00 |
| uvos__ | also you have multitouch gestures | 13:00 |
| ac_laptop | uvos__: we'll I'm not a big fan of touchscreens anyway, I just need to figure out how to configure maemo so it's entirely usable with keys+buttons | 13:01 |
| sicelo | the OS has nothing to do with it. could your cable or port be flakey? | 13:02 |
| uvos__ | hildons keyboard support is really poor | 13:02 |
| uvos__ | like -desktop dosent even let you select stuff with the arrow keys | 13:02 |
| uvos__ | like android dose | 13:02 |
| uvos__ | kinda ironic given that hildons only device was a kbd device | 13:03 |
| uvos__ | while android is mostly used on ts-only devices | 13:03 |
| uvos__ | you can navigate android with just the arrow keys | 13:03 |
| freemangordon | uvos: how's that? ofc you can select with keyboard | 13:05 |
| uvos__ | in hildon-desktop? no | 13:05 |
| freemangordon | yes | 13:05 |
| uvos__ | i cant | 13:05 |
| freemangordon | where? | 13:05 |
| uvos__ | homescreen | 13:05 |
| uvos__ | the app screen | 13:06 |
| freemangordon | umm.... | 13:06 |
| uvos__ | taskt nav | 13:06 |
| uvos__ | *tasknav | 13:06 |
| freemangordon | what you want to select there? | 13:06 |
| freemangordon | there is no text | 13:06 |
| uvos__ | the icons, android lest you select application icons | 13:06 |
| freemangordon | you want to switch between apps with kbd? | 13:06 |
| uvos__ | with the arrow keys and open with enter | 13:06 |
| uvos__ | like desktop linux dose too | 13:06 |
| freemangordon | ah, now I see what "select" means | 13:06 |
| freemangordon | yeah, no such functinality | 13:07 |
| uvos__ | yeah that makes ac_laptops goal of using it with just the kbd not an option | 13:07 |
| ac_laptop | sicelo: Well I assumed leste wasn't able to perform a full charge, since I would never get the green LED with it. With fremantle I get the green LED at the end of a charge. But maybe it was just a calibration issue. Anyway what would be the symptoms of a flakey cable ? I know the original cable I had was not good (couldn't flash u-boot with it) and now I'm using an other one. | 13:07 |
| ac_laptop | uvos__: I know xbindkeys works so I guess I can create a keyboard shortcut to launch a terminal and do everything with the keyboard from there | 13:08 |
| uvos__ | launch termial exists | 13:08 |
| uvos__ | its ctl-x | 13:08 |
| uvos__ | ctrl-shift-x | 13:09 |
| ac_laptop | uvos__: this works on fremantle, I can't do it on a leste fresh install | 13:09 |
| uvos__ | hmm | 13:09 |
| uvos__ | hmmmmm | 13:09 |
| uvos__ | works here | 13:09 |
| uvos__ | another issue for the list | 13:09 |
| uvos__ | https://github.com/maemo-leste/leste-config/blob/master/leste-config-n900/usr/share/hildon-desktop/shortcuts.ini.leste | 13:10 |
| uvos__ | these shortcuts should work | 13:10 |
| uvos__ | if one dosent somethings screwy | 13:10 |
| sicelo | in Fremantle you could switch between taskswotjctrl+bksp | 13:11 |
| sicelo | *with | 13:12 |
| freemangordon | this should work on leste as well | 13:12 |
| uvos__ | you can switch to tasknav (same on leste), but you cant select a window to go to | 13:12 |
| freemangordon | you can | 13:12 |
| freemangordon | but can't remember how :) | 13:12 |
| uvos__ | ah yes you can | 13:13 |
| uvos__ | q is the firs w is second ... | 13:13 |
| uvos__ | cool | 13:13 |
| freemangordon | and you you start typing on desktop, it automatacally serches contacts | 13:14 |
| freemangordon | *searches | 13:14 |
| freemangordon | *and if | 13:14 |
| freemangordon | argh | 13:14 |
| freemangordon | what a typo day | 13:14 |
| uvos__ | but then i cant select a contact with the kbd | 13:14 |
| freemangordon | you should be able to | 13:15 |
| uvos__ | dosent seam to work, neither arrows or tab | 13:15 |
| freemangordon | heh | 13:15 |
| freemangordon | right, you can't | 13:15 |
| freemangordon | that's osso-abook | 13:15 |
| freemangordon | anyway... | 13:15 |
| uvos__ | thats a bit wierd since its just a gtk dialog | 13:15 |
| freemangordon | btw, I just built upstreamxorg | 13:16 |
| freemangordon | not sure how to test it though | 13:16 |
| sicelo | ctrl+bksp hasn't been working for a while ... | 13:18 |
| freemangordon | it has been broken | 13:19 |
| freemangordon | but now it should be functional again | 13:19 |
| freemangordon | unless I have missed something | 13:19 |
| sicelo | will test again. couple of days ago it didn't | 13:20 |
| arno11 | so to resume what's wrong on new img: emmc swap is not activated and this is the biggest problem | 13:22 |
| arno11 | power button works again after reboot | 13:22 |
| freemangordon | uvos: if I make install xorg in /usr/local, would it override my default install? in /usr? | 13:22 |
| arno11 | no troubles with battery | 13:22 |
| uvos__ | freemangordon: well for one thing stuff in /etc will overide /usr/local/share | 13:23 |
| uvos__ | bin is ok since you can just have it later in path | 13:23 |
| uvos__ | im not sure about how its going to load modules | 13:23 |
| uvos__ | it might try /usr first and load the wrong ones | 13:24 |
| freemangordon | yeah | 13:24 |
| uvos__ | sicelo: it should have allways worked on n900 | 13:24 |
| arno11 | apt-worker is activated by default, must be removed (unusable ATM) | 13:24 |
| uvos__ | its "broken" or rather reasigned on d4/bionic | 13:25 |
| freemangordon | hmm, "modeset(G0): using drv /dev/dri/card1" | 13:25 |
| freemangordon | uvos: ^^^ do you remember how to fix that | 13:26 |
| sicelo | ac_laptop: it's curious for sure, but I'm reasonably sure it's not leste's fault. the fuel gauge and charger largely take cate of themselves in mainline | 13:26 |
| uvos__ | freemangordon: hmm i think we figured that one out., i also remember just (re)moving card1 | 13:27 |
| uvos__ | cant quite remember if you can specify this in .conf | 13:28 |
| freemangordon | yes, I think you can, but can't remember how :) | 13:28 |
| freemangordon | kmsdev? | 13:28 |
| uvos__ | maybe no idea | 13:28 |
| freemangordon | yeah, Option "kmsdev" "string" | 13:28 |
| freemangordon | https://man.archlinux.org/man/modesetting.4 | 13:29 |
| uvos__ | Default: /dev/dri/card0 i wonder what i bases choosing card1 on | 13:29 |
| uvos__ | *it | 13:29 |
| freemangordon | no idea | 13:30 |
| arno11 | sicelo: i can confirm, nothing unusual with battery/jauge | 13:31 |
| ac_laptop | sicelo: I'm currently trying a full battery charge with leste booted. But I'm pretty sure the result is going to be the same as before : the green LED indicating full charge will never appear | 13:31 |
| Wizzup | that is correct, that won't happen atm | 13:32 |
| Wizzup | we are no longer carrying a patch from spinal to make that work | 13:33 |
| Wizzup | I don't know why it was dropped | 13:33 |
| ac_laptop | is there a way to "force" calibration ? To tell meamo "this battery is at 100% charge whether you believe it or not" ? | 13:33 |
| sicelo | don't worry about led | 13:33 |
| sicelo | ac_laptop: even being out of calibration won't shutdown thrdevice, so don't worry about it | 13:34 |
| uvos__ | Wizzup: kernel patch? probubly just got lost in the transtion to unified kernel with d4 | 13:36 |
| Wizzup | uvos__: not sure if kernel patch or upower patch | 13:36 |
| uvos__ | ok | 13:36 |
| sicelo | it was upower | 13:37 |
| freemangordon | heh GL_OES_texture_border_clamp required | 13:37 |
| uvos__ | this again | 13:40 |
| freemangordon | yeah | 13:40 |
| freemangordon | but I have a patch around | 13:40 |
| ac_laptop | sicelo: ok well I went back to the device to check the charge, and it's the same thing as usual: the battery icon is showing the charge animation, when I try to use the interface it stops and shows a full battery, while the message "the device is consuming more power than it receives" appears. I check the battery status, it says 100% (with the wrong value of 1794mAh). I guess now I should try | 13:43 |
| ac_laptop | to stress the device until the battery is empty and charge again ? | 13:43 |
| arno11 | ac_laptop: before looking at the battery issue, you must check what is the root cause of "device is consuming more power than it receives" | 13:51 |
| sicelo | show cat /sys/class/power/bq27200-0/uevent | 13:52 |
| sicelo | paste it somewhere | 13:52 |
| arno11 | uvos: freemangordon; btw on a fresh install, 16bpp makes a big difference | 13:54 |
| ac_laptop | sicelo: https://paste.debian.net/1298303/ | 13:55 |
| arno11 | but without transitions ini tweaks, qt5 apps take around 30 sec to open | 13:55 |
| ac_laptop | actually, do I need to do this when it's plugged ? | 13:55 |
| arno11 | only 5-10 sec with zoom and effects deactivated | 13:56 |
| sicelo | heh, your battery is calibrated | 13:56 |
| arno11 | as Wizzup said, probably the lack of qt5 booster | 13:56 |
| sicelo | connect charger, then do cat /sys/class/bq24150a-0/mode | 13:58 |
| sicelo | note I'm writing from memory, somame correction as needed | 13:59 |
| ac_laptop | root@devuan-n900:~# cat /sys/class/power_supply/bq24150a-0/mode | 14:00 |
| ac_laptop | auto (dedicated) | 14:00 |
| sicelo | it sees the charger just fine | 14:01 |
| ac_laptop | out of curiosity what's written on your AC-to-USB charger ? mine says "Output: DC5V.1.3A" | 14:02 |
| Wizzup | arno11 freemangordon - did you figure out if we can use qml in 16bbp or not | 14:02 |
| freemangordon | we can't | 14:03 |
| arno11 | :( | 14:03 |
| freemangordon | I am looking @ modesetting ATM | 14:03 |
| arno11 | :) | 14:03 |
| sicelo | ac_laptop: that's the original charger? | 14:04 |
| freemangordon | no high hopes though | 14:04 |
| freemangordon | it starts, but that's all | 14:05 |
| ac_laptop | sicelo: that's the one that was in the box, anyway | 14:06 |
| ac_laptop | but since the cable was faulty and I had to change it, I want to be sure the specs are ok | 14:07 |
| Wizzup | freemangordon: what would modesetting bring us? working randr on n900? | 14:07 |
| uvos__ | dri3 | 14:07 |
| Wizzup | what does that bring us again? :) | 14:07 |
| uvos__ | surface creation is managed by the client not xorg | 14:07 |
| sicelo | ac_laptop: i think the charger is fine | 14:08 |
| freemangordon | 30 or more % increase in speed | 14:08 |
| freemangordon | ok, upstream ms/glamor starts and xterm is shown | 14:17 |
| freemangordon | however, it hangs the same way as above | 14:17 |
| arno11 | argh | 14:17 |
| freemangordon | hmm | 14:22 |
| freemangordon | actually no | 14:22 |
| arno11 | ? | 14:22 |
| freemangordon | no input drivers are loaded | 14:23 |
| arno11 | uvos: chimaera-devel is still 6.1.48 and chimaera 6.1.9, is it normal ? | 14:29 |
| freemangordon | ok, glamor hangs exactly as berfore | 14:31 |
| freemangordon | Wizzup: ok, can we at least upgrade glibc? | 14:31 |
| freemangordon | if we cannot change the distro? | 14:31 |
| uvos__ | arno11: yeah, altho we should stable | 14:32 |
| uvos__ | arno11: but thats not my department | 14:32 |
| arno11 | ok thx | 14:32 |
| uvos__ | freemangordon: what for @ glibc | 14:32 |
| uvos__ | freemangordon: updateing glibc is probubly a bad idea | 14:32 |
| uvos__ | *we should update stable | 14:32 |
| Wizzup | I mean honestly I'd rather look at the 3d driver situation again later, we're very close to a beta now | 14:37 |
| uvos__ | if ti really wants to open source stuff it might make sense to wait from that perspective too, maybe we can interrogate rs@ti.com about thair intentions with the mesa stuff | 14:40 |
| arno11 | Wizzup: everything seems fine with last image, the only problem is swap | 15:07 |
| arno11 | i'm looking at leste-config-n900.postinst and i think i found what is not working | 15:07 |
| arno11 | addind stuff directly in /etc/fstab without swapon command seems to not work | 15:10 |
| arno11 | i think we rather try to add swapon /dev/mmcblk1p3 command instead of adding stuff in /etc/fstab (swap will be added automatically in fstab with swapon command, i already tried) | 15:15 |
| arno11 | ah and default swappiness value seems to high for N900 | 15:22 |
| freemangordon | uvos: new sgx blobs require 2.34 | 15:23 |
| freemangordon | Wizzup: ok, maybe I can upgrade my device only | 15:24 |
| freemangordon | to see if it gets any better with the new blobs in regards to ms/glamor and in general | 15:24 |
| arno11 | *too | 15:27 |
| ac_laptop | I'm genuinely interested to try a D4 now | 15:34 |
| ac_laptop | There doesn't seem to be many offers on the internet though, it's not liek the n900 | 15:34 |
| ac_laptop | like* | 15:35 |
| uvos__ | it was only ever sold in the united states of america | 15:40 |
| uvos__ | so anywhere else you need to import it | 15:40 |
| uvos__ | also they have become sortof rare at this point | 15:40 |
| uvos__ | it used to be that you could just pick one up for 25 euros | 15:40 |
| uvos__ | but the existance of leste drew the pool down pretty quick | 15:41 |
| ac_laptop | I see there are some verizon-simlocked online | 15:58 |
| ac_laptop | I guess I can unlock them but I will need to have a google account and boot under android first, right ? | 15:59 |
| uvos | ac_laptop: evey single droid 4 is simlocked to verizon | 16:15 |
| uvos | ac_laptop: but only in the us, they are unlocked everywhere else | 16:15 |
| uvos | you dont have to interact with android at all | 16:15 |
| uvos | besides android dosent need a google account | 16:15 |
| ac_laptop | uvos: If I buy a droid 4 from the U.S. sold as "verizon-locked", what will happen ? | 16:18 |
| Wizzup | arno11: we do not want to swapon in image builder | 16:18 |
| arno11 | Wizzup: ok so maybe use an init script ? | 16:31 |
| arno11 | for swap and swappiess | 16:31 |
| arno11 | *swappiness | 16:31 |
| arno11 | btw the weird thing is that boot msg says swap is activated but in fact not | 16:38 |
| arno11 | (on first boot) | 16:38 |
| uvos | ac_laptop: as i said all d4 are verizon-locked but they are locked in the us only, as soon as they cross the border they are unlocked and will connect to anny carrier | 17:24 |
| ac_laptop | uvos: ok, that's good then :) | 17:29 |
| ac_laptop | uvos: but then imagine I go to the US with it, that means I need to swap the SIM for a verizon one I suppose ? | 17:30 |
| uvos | ac_laptop: no its totaly useless in us | 17:31 |
| uvos | ac_laptop: verizon refuses to activate phones of this age | 17:32 |
| uvos | if you want a simmular device with no simlock you can buy a droid bionic, its also fully supported by leste, and this one has no sim lock, but also no keyboard | 17:32 |
| ac_laptop | uvos: I'll test a d4 along with my n900 and see the one I'm the most comfortable with. :) If it's the d4 I'll just have to remember never to travel to the US with it | 17:37 |
| uvos | i dont think you will be able to use the n900 for another minute after using the d4 given the delta in performance :P | 17:38 |
| ac_laptop | uvos: yeah, and from the pictures the keyboard seems more comfortable too :) | 17:39 |
| Wizzup | arno11: are you sure that /etc/fstab with swap doesn't auto mount swap? | 17:41 |
| Wizzup | I'm quite sure that it should | 17:41 |
| ac_laptop | 38 keys vs 54 | 17:42 |
| ac_laptop | can't wait :) | 17:42 |
| arno11 | Wizzup: i can certify it doesn't work :( | 17:45 |
| arno11 | uvos: don't forget that n900 is perhaps 300% faster with few tweaks ;) droid is faster, that's a fact but n900 is now usable with lot of fun | 17:48 |
| arno11 | and enough if a user or dev don't need modern browsing | 17:49 |
| bencoh | (and n900's keyboard actually feels better imho, just with heavy remapping) | 17:50 |
| Wizzup | arno11: ok, let me try to debug that a bit further | 17:52 |
| Wizzup | maybe we just need to add 'auto' | 17:52 |
| uvos | i strongly disagree on the keyboard, its just so flat and crampt on the n900, other wise sure it can be fun to make things work with limited resources | 17:53 |
| uvos | but the keyboard is allways a personal preferance/what you used more kind of thing | 17:53 |
| arno11 | Wizzup: ok :) | 17:53 |
| bencoh | uvos: true | 17:54 |
| Wizzup | arno11: maybe there is some swap init script that is not started | 17:58 |
| Wizzup | arno11: I did find this somewhere: | 17:59 |
| Wizzup | > | 17:59 |
| Wizzup | Exchange auto with sw in the fstab line. This tells the system to activate it using swapon after booting. | 17:59 |
| Wizzup | I'll toy with this later today | 18:00 |
| arno11 | ok interesting | 18:00 |
| ac_laptop | uvos: n900 keyboard is not the worst keyboard but the keys are so small that it's not very comfortable to press two close ones at the same time | 18:01 |
| ac_laptop | Shift-Ctrl-X in particular is not fun to do on the n900 :) | 18:01 |
| arno11 | ac_laptop: true | 18:01 |
| arno11 | and keys are weak | 18:02 |
| ac_laptop | apart from that it's ok (I have big fingers, I mostly use the nails of my thumbs on this keyboard) | 18:03 |
| ac_laptop | bencoh: which remapping do you have ? do you switch between mappings or not ? | 18:27 |
| bencoh | I have a custom xkb map for n900/fremantle from almost 10y ago | 18:28 |
| ac_laptop | bencoh: do you know if there's a way to simulate the alt key ? | 18:44 |
| bencoh | not sure what you mean by 'alt key' | 18:47 |
| ac_laptop | bencoh: an additional modifier key | 19:00 |
| ac_laptop | Meta key in emacs | 19:00 |
| bencoh | with xkb you could set any key as a modifier, so ... | 19:23 |
| bencoh | (pretty much at leasT) | 19:23 |
| arno11 | Wizzup: i had a look and indeed we must use 'sw' instead of 'defaults' apparently | 20:25 |
| Wizzup | does that work for you? | 20:27 |
| arno11 | so the correct line to add seems to be '/dev/mmcblk1p3 swap swap sw 0 0' iiuc | 20:27 |
| arno11 | i'll try a bit later tonight and let you know | 20:28 |
| Wizzup | ty | 20:29 |
| arno11 | no probs | 20:29 |
| sicelo | ac_laptop: i think Fn key is R_Alt | 20:35 |
| arno11 | Wizzup: i can't reproduce the error: defaults or sw or even wizzup work in fstab. without rebuilding with new parameter (sw), i don't know what to do to confirm if it's really the problem | 22:42 |
| arno11 | and apparently sw option is not working on first boot with every distros | 22:43 |
| Wizzup | right, so it could be a first boot only problem? | 22:45 |
| arno11 | yes definitely | 22:46 |
| arno11 | so i think if you can rebuild with the sw parameter, i can reburn the img and test | 22:49 |
| Wizzup | I think it seems like we have another problem not related to fstab | 22:58 |
| arno11 | ah | 22:58 |
| arno11 | what kind of issue ? | 23:00 |
| Wizzup | well, you said we have a problem on first boot only :) | 23:00 |
| arno11 | yes lol because maybe the sw option is missing in postinstall fstab stuff | 23:01 |
| Wizzup | I would like to avoid rc.S swapon -a style hacks if we can, so it might take a bit more debugging | 23:02 |
| Wizzup | it could be a devuan init script or so | 23:02 |
| arno11 | ok | 23:03 |
| arno11 | maybe there is an easier way: if we remove the expand card message, deactivate nokia-modem and apt-worker on boot, i think n900 can run easely without swap on first boot (just an idea) | 23:07 |
| arno11 | (i'm currently running n900 with almost no swap and it works well, rarely need more than 200megs of swap) | 23:10 |
| arno11 | and then users/devs could choose the best option regarding swap | 23:11 |
| arno11 | again, just an idea | 23:12 |
| Wizzup | I think we probably just want some swap working, that's easier to solve than not running certain things on boot | 23:13 |
| arno11 | hope it will be easier to solve | 23:19 |
| arno11 | *200megs of ram | 23:20 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!