14:01 | Lo-lan-do | Hi all :-) |
14:03 | Lo-lan-do | I've been fighting to get wlan working on my Cubox-i4pro. |
14:04 | Lo-lan-do | I'm using the stock Debian Jessie kernel, have installed all firmware packages I could find, but no luck so far. |
14:05 | Lo-lan-do | wlan0 persists in not appearing, and dmesg doesn't say anything useful. |
14:09 | curlymo | does the same thing happen in e.g. XBian or OpenElec? |
14:10 | suihkulokki | worth specifying that debian jessie comes with 3.16 mainline kernel |
14:10 | Lo-lan-do | I used a preinstalled image at some point before switching to pure Debian, and it did work. |
14:13 | Lo-lan-do | Would the patches work on 3.16? |
14:14 | curlymo | XBian runs on 3.14.27+ |
14:20 | jnettlet | Lo-lan-do, Upstream is fine for experienced hackers but please use the SolidRun kernel for the best support. |
14:36 | Lo-lan-do | That's... sounding kind of backwards, isn't it? |
14:37 | jnettlet | Why is that backwards. Don't you believe that the company providing the hardware is going to put the best effort into providing the best support for their product? |
14:39 | jnettlet | Not to mention that 3.16 is actually missing a lot of patches that hadn't even been accepted to be merged yet, so the 3.16 support is way behind 3.18 which still requires patches that haven't been accepted yet. |
14:40 | jnettlet | it takes a long time to all the needed patches upstream. Much longer than a products development life-cycle. That is why LTS kernels came about |
14:41 | suihkulokki | jnettlet: is the missing patches for 3.16/3.18 documented somewhere? |
14:41 | suihkulokki | if they are not too intrusive, there is a chance of convincing them into debian jessie release |
14:42 | jnettlet | suihkulokki, rmk maintains them in his personal repository and patch release sets. |
14:42 | jnettlet | nope, debian has told us to go screw pretty much :) |
14:42 | jnettlet | that is why we are building our own apt repository to better support our hardware on Debian |
14:43 | jnettlet | We have wheezy packages built and are working on jessie now |
14:43 | jnettlet | including accelerated chromium with webgl and HTML5 accelerated decoding support |
14:46 | Lo-lan-do | Ah. So there's an apt repo with SolidRun kernel packages somewhere? |
14:46 | Lo-lan-do | That sounds promising :-) |
14:46 | jnettlet | for wheezy |
14:47 | jnettlet | we are still working on jessie builds |
14:47 | jnettlet | I think it is announced in the forums |
14:49 | jnettlet | We tried for a while to get better distro support, and in the end decided we had to do our own thing in order to not make compromises and a bunch of needless extra work. |
15:02 | suihkulokki | jnettlet: sad to hear but I totally understand |
15:04 | jnettlet | We are trying to bring the best software experience to the community. We will support our own packages while we continue to upstream and develop full OSS solutions. |
15:06 | palatin | does it apply to cubox v1 also ? (I worry about security issues related to kernel growing old, 3.6.9 atm) |
15:09 | suihkulokki | palatin: I'm running cubox v1 a pure mainline kernel (3.17-rc6, forgot to update for a while) |
15:10 | suihkulokki | palatin: however I use as headless machine atm. Graphics and multimedia probably wouldn't work on mainline really well |
15:11 | palatin | I am also headless, using it as a router/firewall/server |
15:11 | palatin | care to share your .config ? |
15:14 | jnettlet | palatin, we are working on fixing up support for the original Cubox in the upstream kernel |
15:14 | jnettlet | there has been some breakage that needs to be tracked down. |
15:14 | jnettlet | suihkulokki, don't update past that :) |
15:14 | jnettlet | the breakage happened sometime in 3.17-3.18 |
15:15 | jnettlet | well unless you want to be nice enough to git bisect the breakage :P |
15:15 | suihkulokki | jnettlet: i can bisect but not tonight |
15:16 | jnettlet | but for better graphics support on the original Cubox we are working on it through etnaviv. Unfortunately the Armada chip in it only has iWMMXt support so newer libGAL libs can't be used. |
15:16 | suihkulokki | I knew things would break once I stopped updating :P |
15:16 | palatin | jnettlet: ok I was worried, from a user point of view v1 support seems pretty much abandoned |
15:17 | jnettlet | palatin, well things tailed off since most support went upstream. It is the primary testing platform for the Xorg work that is ongoing with etnaviv |
15:17 | jnettlet | well for rmk at least. |
15:18 | suihkulokki | palatin: http://kos.to/cubox/cubox-3.13-rc7.config |
15:19 | palatin | suihkulokki: thanks ! |
15:20 | suihkulokki | jnettlet: is there some mailing list thread etc about the breakage or will it be obvious right away? |
15:20 | jnettlet | suihkulokki, I believe it won't boot |
15:20 | jnettlet | so bad |
15:20 | jnettlet | most likely some bad device-tree entry |
15:36 | R0nd | jnettlet: any updates on that kodi problem where some files wouldn't play any video? I'm building the latest source right now |
15:41 | rabeeh | R0nd: which one is that? |
15:53 | R0nd | rabeeh: kodi wouldn't render any video for some files, it was inconclusive whether it depends on codec or something else. I think jnettlet said something about vpu not doing anything after looking at my logs |
15:54 | jnettlet | R0nd, sorry, yes you weren't building Kodi properly to use the MX6 codec which basically means to use the VPU to decode video. |
15:54 | jnettlet | everything was trying to be decoded through software and couldn't keep up. |
15:55 | R0nd | jnettlet: do I need to reconfigure? |
15:55 | topi` | received a batch of Carrier Pro's but how do I find out if they are rev.3.5 or the older one? |
15:56 | topi` | I soldered a SIM holder to the pads and the sim doesn't seem to work, so I'm inclined to believe it's the older, bugged one |
15:58 | jnettlet | R0nd, make sure you have --enable-codec=imxvpu |
16:01 | topi` | is there any text or marking on the Carrier board that indicates the revision? |
16:02 | rabeeh | top`: i'll PM |
16:02 | R0nd | jnettlet: yes, I have that option. I can certainly say that at least kodi's gui is rendered on hw |
16:03 | jnettlet | yes, that is through the 3D engine. The VPU is the hardware video decoder |
16:04 | rabeeh | rev of the HummingBoard is beneath the MicroSOM ! |
16:05 | R0nd | jnettlet: here's the list of options I'm using anyway --disable-vdpau --disable-vaapi --enable-gles --disable-pulse --disable-projectm --enable-codec=imxvpu --disable-openmax --disable-x11 --disable-xrandr --disable-gl --disable-sdl |
16:13 | jnettlet | R0nd, well give it a build and test things out. Then send me logs again if there are problems. |
17:20 | R0nd | jnettlet: same deal, here's the log http://pastebin.com/KLQhtmSD |
17:23 | jnettlet | R0nd, 19:17:47 T:1757410368 DEBUG: FactoryCodec - Video: iMX-h264 - Opened 19:17:47 T:1757410368 NOTICE: Creating video thread 19:17:47 T:1757410368 ERROR: Get - failed to get stream |
17:23 | jnettlet | is this being played from the local drive? |
17:24 | R0nd | yes, local esata hdd |
17:24 | R0nd | I can play at least one other file from the same drive without problem |
17:27 | jnettlet | R0nd, I bet I know what is going on. How much memory are you allocating to CMA? |
17:28 | R0nd | no idea |
17:28 | jnettlet | it should be at the top of dmesg |
17:29 | jnettlet | within the first 20 lines or so |
17:32 | R0nd | I'm afraid it's not |
17:32 | R0nd | dmesg output http://pastebin.com/PD3YcPHs |
17:34 | jnettlet | lol. |
17:35 | jnettlet | disable the ev debug support |
17:40 | R0nd | jnettlet: cma: CMA: reserved 256 MiB at 6e000000 |
17:40 | jnettlet | okay I was wrong |
17:40 | jnettlet | damn. |
17:40 | jnettlet | can you give me full dmesg please? |
17:41 | R0nd | http://pastebin.com/QEx8tDfk |
17:51 | jnettlet | Is that after playing a failed video? |
17:51 | R0nd | yes |
17:51 | R0nd | no, wait |
17:52 | R0nd | the dmesg is not |
17:57 | R0nd | here's one after playing a video http://pastebin.com/BjBD4pdg |
18:00 | jnettlet | nothing. very strange |
18:04 | jnettlet | R0nd, do you have the debug option enabled under system settings for Kodi? |
18:09 | R0nd | jnettlet: here's a debug log http://pastebin.com/psUZUWUr |
21:57 | rabeeh | http://download.solid-run.com/pub/solidrun/HummingBoard/rev-3.5/HummingBoard-PCB-Expedition-rev-3.5.zip |
21:58 | rabeeh | HummingBoard layout in Expedition format (for the interested) |
21:58 | rabeeh | and - http://download.solid-run.com/pub/solidrun/SR-uSOM-mx6/AN2/SR-uSOM-AN2-Suspend-to-memory-power-consumption.pdf |
21:58 | rabeeh | suspend to memory power measurements |
22:10 | PsilonuX | Hi, I am the one with the microsd popping out.. Here is some proof: http://youtu.be/U7EJCsB348Q |
22:11 | PsilonuX | If I am lucky it doesnt pop out, but jumps out more then 10cm, no joke |
22:12 | PsilonuX | its like spring loaded.. |
22:12 | PsilonuX | but with a bad lock |
22:13 | PsilonuX | But for the time it works, it looks like an awesome device |
22:21 | kgp | Hi rabeeh! |
22:22 | kgp | Just to say hellp ;) |
22:22 | kgp | hello |
22:22 | Artox | PsilonuX: lol'd |
22:22 | Artox | greta invention |
22:22 | Artox | great* |
22:23 | PsilonuX | I spilled some magic fairy dust over it! lol |
22:26 | kgp | Personnaly, I have a long nail now :) it's another solution for the "spring problem" ;) |
22:31 | Artox | are there any debian(/apt) experts around who can tell me when this error can happen and what it means?: E: GPG error: http://repo.maltegrosse.de ./ Release: The following signatures were invalid: NODATA 1 NODATA |