| 00:18 | aquarat | if I reflash u-boot manually will it retain the environment variables ? |
| 00:18 | aquarat | oh, nvm "save" |
| 00:32 | aquarat | yes! the installer is booting |
| 00:33 | aquarat | it kept complaing about a bad architecture when it loaded the boot.scr.serial file |
| 00:33 | aquarat | so I tried manually loading the kernel... but that failed too because fatload can't seem to handle files outside the root of the drive |
| 06:10 | jnettlet | so the bottom usb port is working in u-boot but not the top. :-\ |
| 06:16 | dbsx | jnettlet: on your new uboot on the board? I could never get anything but the top one to boot on the old cubox and similar issues on marvell dreamplugs. |
| 06:17 | jnettlet | dbsx, on the new uboot on the carrier-1 board. yep |
| 06:17 | jnettlet | maybe a u-boot limitation. I will add it to my list. |
| 06:17 | jnettlet | but with a thumb drive in the bottom port usbboot is possible |
| 06:18 | jnettlet | useful for recovery. |
| 06:18 | dbsx | If you ask rabeeh, he (I think) will agree that denx USB is a little shaky. |
| 06:20 | jnettlet | if I get some free time I will compare it to openfirmware's usb support. Mitch has put a lot of work into making that very stable. |
| 06:23 | dbsx | I had a brief look at you uboot git. Looks great. But how does the environment on mmc work when you want to update it? |
| 06:24 | jnettlet | when you want to update u-boot? |
| 06:24 | dbsx | Just the environment |
| 06:25 | jnettlet | you can replace any of the env variables with your boot.scr |
| 06:26 | jnettlet | oh you mean like saveenv which normally writes to flash. |
| 06:26 | dbsx | Or did I misread the code? It looks like bot uboot and its env are on disk. Which means it would be difficult to update. |
| 06:26 | dbsx | ok |
| 06:26 | dbsx | boot.scr is the way out then |
| 06:27 | jnettlet | yep. I mimicked what most the other commercial IMX devices are doing to give a good general working environment. User customization can be done compiling your own or using a boot.scr |
| 06:27 | dbsx | Thanks. I guess I was asking if saveenv was possible to mmc. |
| 06:28 | jnettlet | hmmm actually. saveenv seems to try to write to the MMC card but reports it is locked. |
| 06:29 | jnettlet | popping it out and in is a little more succesfull |
| 06:29 | jnettlet | C1 U-Boot > saveenv |
| 06:29 | jnettlet | Saving Environment to MMC... |
| 06:29 | jnettlet | Writing to MMC(0)... failed |
| 06:30 | jnettlet | but then you would need the mkImage code in u-boot....can see how that could work, but not sure how practical |
| 06:31 | dbsx | You could save some space and remove saveenv. Building your own uboot is a reasonable way out. |
| 06:31 | dbsx | even a dd to the mmc to install a new environment is conceivable |
| 06:34 | jnettlet | oh I just need to include the write modules for fat and ext2 |
| 06:34 | dbsx | The other question is whether mtdparts mean anything when using mmc? |
| 06:34 | dbsx | great! |
| 06:35 | dbsx | x-over typing |
| 06:38 | jnettlet | well it is something to think about. I think for now I am going to leave it needing a custom boot.scr. Time to focus on kernel work :-) |
| 06:38 | dbsx | Good choice. u-boot is hard labour |
| 06:39 | jnettlet | I only tackled this problem to get network booting sorted to make kernel dev work easier. |
| 06:40 | jnettlet | u-boot isn't so bad. Just not very rewarding. Ideally your code should run and be done faster than you can blink your eyes :-) |
| 06:40 | jnettlet | except unless you are developing on the platform of course. |
| 06:42 | dbsx | Thanks for your efforts. |
| 06:43 | jnettlet | np |
| 06:47 | dbsx | jnettlet bows to thunderous applause |
| 06:54 | jnettle | 06:54 * jnettlet thinks the thunderous applause will be when I release the new graphics driver kernel today or tomorrow |
| 07:00 | dbsx | good luck |
| 11:18 | jnettlet | dv_, is your meta for the cubox-i ready for consumption yet? |
| 11:22 | dv_ | you mean the OE meta-cubox layer? |
| 11:23 | dv_ | otavio will add support for the carrier one and the cubox-i to meta-fsl-arm |
| 11:24 | dv_ | I sent him my modifications to support the carrier one. they are hacks just to get something usable. |
| 11:26 | jnettlet | dv_, ah okay you are using meta-fsl-arm. Are those patches anywhere that I can grab them? |
| 11:28 | dv_ | I can forward the patches I sent to otavio |
| 11:29 | jnettlet | thanks |
| 11:33 | dv_ | i will not give you something that automatically boots the kernel |
| 11:33 | dv_ | so you have to type in u-boot cmdlines |
| 11:34 | dv_ | I mainly did it to have a working system for my gstreamer work |
| 11:34 | jnettlet | dv_, I will point it at my uboot repo |
| 11:34 | dv_ | oh, and dont use the .sdcard image |
| 11:34 | jnettlet | okay |
| 11:34 | dv_ | its not adapted for the c1 yet |
| 11:38 | dv_ | also, whats this I hear about an upcoming kernel module release? :) |
| 11:39 | jnettlet | yep finally getting back to that. |
| 11:40 | jnettlet | working on it right now. |
| 11:42 | jnettlet | looking like tomorrow night I should have enough done to release something |
| 11:58 | _rmk_ | jnettlet: I haven't worked out if the top or bottom is the OTG port, but for the OTG port you need to ensure that the ID pin is held low |
| 11:59 | jnettlet | _rmk_, the micro-usb is only a power source? I thought that was the OTG. |
| 11:59 | jnettlet | Figured it was just re-purposed as there is no need for OTG when you have two full sized ports |
| 12:00 | _rmk_ | I don't think that port is wired up on its USB pins |
| 12:00 | _rmk_ | afaik its just used for power |
| 12:02 | jnettlet | Well this would be the only USB A-port that I know of that supports otg. I guess that means we could turn that into a serial port if need be. |
| 12:03 | _rmk_ | it doesn't support otg because its only 4 pin - looking at Rabeeh's patches... |
| 12:04 | _rmk_ | + /* |
| 12:04 | _rmk_ | + * Originally MX6Q_PAD_GPIO_1__USBOTG_ID, but we want it to be |
| 12:04 | _rmk_ | + * pulled down ID pin for fixed host connection. |
| 12:04 | _rmk_ | + */ |
| 12:04 | _rmk_ | that's to turn it into a host-only port |
| 12:06 | _rmk_ | yep, just traced through the sysfs stuff, the IMX6 OTG USB is used as one of the USB host ports |
| 12:07 | _rmk_ | and the iomux settings are used to ensure its always in host mode |
| 12:07 | _rmk_ | and yes, it's the top USB port |
| 12:09 | _rmk_ | because when I plug my pendrive into it, I get this path appear: |
| 12:09 | _rmk_ | /sys/devices/soc.0/2100000.aips-bus/2184000.usb/ci_hdrc.0/usb1/1-0\:1.0/port1/device/1-1\:1.0/ |
| 12:09 | _rmk_ | and 0x2184000 is the IMX6 OTG port |
| 12:14 | jnettlet | okay. I should add that to u-boot then. |
| 12:19 | _rmk | 12:19 * _rmk_ rotfls as he reads about someone with an IMX6Q which times out on the network when both the SATA and ether are active |
| 12:19 | _rmk_ | connected a 40A PSU to the board, so it can't be a power supply problem! |
| 12:19 | _rmk_ | *doh* ! |
| 12:19 | jnettlet | ouch |
| 12:21 | _rmk_ | those of us who understand electronics and the design of these things will realise that there's on-board regulators between the supply input and the CPU/SDRAM |
| 12:21 | _rmk_ | and these regulators can cause problems |
| 12:22 | _rmk_ | I almost thought about posting "maybe you should try a different power station?" |
| 12:24 | jnettlet | or a different developer :-) |
| 12:24 | jnettlet | hmm usb otg is going to take more time than I have right now. Will put it on the list for later. |
| 12:25 | jnettlet | Don't think it will hold anyone up right now |
| 12:29 | jnettle | 12:29 * jnettlet wonders what fun toys rabeeh will have when he appears back online. |
| 13:58 | otavio | jnettlet: in case you're working i nthe yocto support I can help |
| 14:10 | jnettlet | otavio, yocto support for which? |
| 14:10 | jnettlet | what is a yocto anyways? |
| 14:15 | dv_ | jnettlet: http://yoctoproject.org/ |
| 14:15 | dv_ | yocto is sort of an umbrella project to coordinate OpenEmbedded, bitbake etc. |
| 14:15 | jnettlet | dv_, yeah I know it is the distro. But I don't know what an actual yocto is. |
| 14:15 | dv_ | there is no "yocto" |
| 14:16 | jnettlet | maybe an octagonal yoyo |
| 14:16 | dv_ | yocto is actually an SI prefix |
| 14:16 | wumpus | there is, but we're not allowed to talk about that :) |
| 14:18 | jnettle | 14:18 * jnettlet should probably stop looking into the dark underbelly of the yocto secret society |
| 14:18 | dv_ | it can be confusing, yes |
| 14:19 | dv_ | since in theory, you could use OE directly, without the yocto extras |
| 14:19 | dv_ | and you dont really download "yocto", you download "poky", the official yocto distro |
| 14:22 | otavio | heh a mess in fact |
| 14:22 | otavio | lol |
| 14:24 | dv_ | I remember it this way: yocto = umbrella project to coordinate development of the component projects (bitbake, OE, poky) to ensure they play well together ; OE = bunch of bitbake recipes for building ; poky = OE distribution supported by Yocto, containing a preconfigured bitbake, a copy of OE-core, and some extra yocto OE recipes in additional OE layers |
| 15:10 | MikeSeth | argh, a month and a half to wait, argh |
| 15:10 | MikeSeth | I am going to die |
| 16:21 | jnettlet | _rmk_, did you attempt to get the imx KMS driver brought up to mainline yet? |
| 16:59 | rabeeh | jnettlet: ping |
| 16:59 | rabeeh | i'm back online |
| 17:00 | rabeeh | _rmk_ / jnettlet : with regards the ID pin; I can force it to GND on the board to get rid of the hack to force it to host |
| 17:00 | rabeeh | what do you think? |
| 17:07 | jnettlet | rabeeh, welcome back |
| 17:07 | rabeeh | jnettlet: thanks |
| 17:08 | rabeeh | sorry for my absence; we are spinning the boards for final production |
| 17:08 | jnettlet | rabeeh, I expected it no problem. |
| 17:08 | rabeeh | so it's quite a huge effort of reviews |
| 17:08 | jnettlet | I think forcing that pin to ground would make sense. It would make the configuration simpler and avoid some confusion as the port will never really act as an OTG port |
| 17:09 | rabeeh | ok |
| 17:10 | rabeeh | jnettlet: what happened to your board? i'v seen it had some issues? |
| 17:10 | rabeeh | you mentioned the fuse? |
| 17:11 | jnettlet | rabeeh, yeah I blew the fuse. |
| 17:11 | jnettlet | I have removed it for now. Didn't know what to replace it with. |
| 17:12 | rabeeh | how did you blow it? |
| 17:12 | rabeeh | it's a 5A fuse for protection |
| 17:12 | rabeeh | do you think it's our fault with anything? |
| 17:13 | dv_ | maybe a power surge |
| 17:13 | dv_ | ? |
| 17:13 | rabeeh | i doubt this is the case |
| 17:13 | rabeeh | power surge will kill the power supply first |
| 17:14 | jnettlet | rabeeh, I wasn't doing too much. I had just re-powered the board. |
| 17:15 | jnettlet | maybe I just got a bum fuse |
| 17:15 | jnettlet | definitely nothing happened that should have blown a 5 amp fuse. |
| 17:15 | jnettlet | and nothing else on the board was damaged. |
| 17:16 | rabeeh | there is a TVS diode that will absorb any voltage that is beyond 6V |
| 17:16 | jnettlet | and it actually didn't even have a burnt smell to it. First thing I did was sniff around the board |
| 17:16 | rabeeh | oh |
| 17:16 | rabeeh | so how did you fix the board then? |
| 17:16 | jnettlet | I removed it. |
| 17:17 | jnettlet | yes I know that is a bad idea |
| 17:17 | rabeeh | no; it's not |
| 17:17 | rabeeh | you want a working board :) |
| 17:17 | rabeeh | which power supply are you using? |
| 17:17 | jnettlet | that was my thought |
| 17:17 | rabeeh | i mean the voltage and the max current |
| 17:17 | jnettlet | I had a 5volt 2amp from something. I swapped that out for my 5volt 1.5 samsung which should be better quality |
| 17:18 | rabeeh | one last thing; do still have the fuse? can you check it's resistance? |
| 17:19 | rabeeh | (i.e. if it's really blown or maybe not soldered good enough on the board) |
| 17:19 | rabeeh | jnettlet: sorry for nagging on this but i just want to check if we have an issue we need to fix or not |
| 17:20 | rabeeh | and there will be a next test; checking the output voltage of the older power supply under same load if it over voltage or not |
| 17:20 | jnettlet | rabeeh, I checked it on the board and it was blown |
| 17:21 | rabeeh | ok |
| 17:21 | rabeeh | if you board dies please ping asap |
| 17:21 | jnettlet | yep so far it is running along happily. |
| 17:32 | jnettlet | rabeeh, is there any more functionality you were thinking would be useful in u-boot? |
| 17:32 | rabeeh | is gig working? |
| 17:32 | jnettlet | yep |
| 17:32 | rabeeh | :) |
| 17:32 | rabeeh | jnettlet: WOW |
| 17:32 | rabeeh | so fast ! |
| 17:33 | rabeeh | and i'v seen the logo on HDMI :) |
| 17:33 | jnettlet | well really thank the boundary guys. They have put a lot of functionality into u-boot for the imx6 |
| 17:33 | rabeeh | oh; is there a real prompt available on the HDMI port? |
| 17:34 | jnettlet | I have to look into that. I have the console mux enabled and am getting hdmi output, but don't get a CLI. |
| 17:34 | jnettlet | usb is enabled and works but I am not enabling it by default as it is a bit slow. |
| 17:34 | rabeeh | slow? |
| 17:35 | jnettlet | just scanning the bus |
| 17:35 | rabeeh | ok; it's probably enabling some time for USB hard drives to be spun before being scanned |
| 17:35 | rabeeh | (i hope 'spun' is the right English word) |
| 17:36 | jnettlet | A lot of functionality is in a "working" state and I figured we could figure out how to implement it after more people played with their hardware and asked for us to do something with it. |
| 17:36 | jnettlet | yep spun is right |
| 18:25 | dv_ | jnettlet: do you know of http://rcn-ee.net/deb/saucy-armhf/v3.11.3-armv7-x14/ ? |
| 18:26 | dv_ | it contains something for imx drm (hdmi specific) |
| 18:26 | jnettlet | interesting. I will take a look |
| 18:27 | dv_ | oh wait |
| 18:28 | dv_ | in its config, there is CONFIG_DRM_IMX=y |
| 18:28 | dv_ | but I cant find a source |
| 18:30 | jnettlet | dv_, that is probably for the KMS driver that is floating around mailing lists |
| 18:30 | jnettlet | It might have been accepted to linux-next still poking around for the latest version. |
| 19:10 | _rmk_ | the IMX DRM is in drivers/staging/ |
| 19:10 | jnettlet | excellent |
| 19:51 | _rmk_ | well, IMX DRM has been in staging for over a year now, I suspect gregkh will get grumpy if we try and add HDMI stuff to it |
| 19:53 | dv_ | why? |
| 19:53 | dv_ | is staged stuff supposed to not be touched? |
| 19:54 | _rmk_ | the point of staging is to clean it up so that it can move out of staging into the proper place |
| 19:54 | _rmk_ | it isn't there to have stuff pushed into mainline and then forgotten |
| 19:56 | _rmk_ | one of the TODO points is to get David Airlie to look at it - I suspect that's not going to happen all the time it exists in staging, because its off everyone's radar |
| 19:56 | _rmk_ | in other news... I now have a profile pic^w^wselfie... I took today :) |
| 19:57 | _rmk_ | at about 900ft above the ground :) |
| 20:03 | rabeeh | _rmk_: we can send hardware to David if needed |
| 20:28 | jnettlet | _rmk_, great selfie. Didn't know you were into gliders |
| 20:43 | dv_ | otavio: how far along is the c1 integration ? if you guys wont have time in the foreseeable future, I could try to clean up what I have |
| 20:43 | dv_ | (assuming I find the time) |
| 20:55 | jnettlet | dv_, do it! do it! |
| 20:56 | dv_ | :)= |
| 20:57 | dv_ | just want to make sure there are no duplicate efforrts |
| 20:57 | dv_ | damnit, I cant type today |
| 20:57 | jnettle | 20:57 * jnettlet hates it when he has a can't type day. |
| 20:59 | jnettlet | hmmm an 11" chromebook running ARM |
| 20:59 | jnettlet | I still think the Samsung model looks better. |
| 21:00 | jnettlet | We need the Cubook-i |
| 21:01 | dv_ | yay |
| 21:25 | aquarat | so the default ubuntu installation has no dhclient ? |
| 21:26 | aquarat | why does my cubox complain of an invalid architecture when trying to load the cubox installer ? |
| 21:33 | otavio | dv_: if you can, better; I am on 3.10 integration |
| 21:38 | dv_ | otavio: linux-imx 3.10 or linux-fslc 3.10? |
| 21:38 | otavio | dv_: linux-imx; the linux-fslc is going to 3.12 ;-) |
| 21:39 | dv_ | jnettlet: how would your changes fit into this? |
| 21:40 | dv_ | iirc, linux-fslc is mainline + a few extras, without the VPU and GPU drivers? |
| 22:01 | _rmk_ | jnettlet: sorry, was on a (long) phone call with a friend at the CAA :) |
| 22:02 | _rmk_ | I was going to upload a couple of other photos from today too |
| 22:05 | jnettlet | dv_, 3.12 is fine. I am working on 3.10 with necessary backported patches. |
| 22:06 | dv_ | perhaps your stuff can then be used with linux-imx 3.10 |
| 22:15 | dv_ | also, any news on the 640x480 problem? |
| 22:17 | jnettlet | haven't gotten to it yet. u-boot has dominated my Cubox hacking time. |
| 22:18 | dv_ | speaking of it, is it uploaded somewhere? I could use it for my OE carrier-one recipes. |
| 22:18 | dv_ | if not, I just use rabeeh's modified 2009.08 for now |
| 22:19 | jnettlet | dv_, yeah it is under my github account |
| 22:19 | dv_ | oh, cool |
| 22:19 | jnettlet | https://github.com/linux4kix/u-boot |
| 22:19 | jnettlet | imx6 branch |
| 22:19 | jnettlet | which is the only one up there right now |
| 22:20 | dv_ | anything I should be careful about? |
| 22:22 | jnettlet | dv_, nope it should all just work. There are different build instructions because it will build the new .imx image. |
| 22:23 | jnettlet | dv_, https://plus.google.com/112696520735663897193/posts/Uf5qHmmnTQp |
| 22:23 | dv_ | I'll just reuse and adapt the existing uboot 2013 recipes |
| 22:25 | jnettlet | should be fine. |
| 23:01 | dv_ | otavio: actually, should the carrier one be added to meta-fsl-arm or meta-fsl-arm-extra ? |
| 23:01 | otavio | dv_: extra |
| 23:22 | _rmk_ | jnettlet: there you go, a couple more photos for you :) |
| 23:27 | otavio | jnettlet: how far is u-boot support? |
| 23:29 | _rmk_ | otavio: the version I ran from jnettlet yesterday has networking, mmc, one usb port working, and HDMI output at a fixed resolution of 1024x768 @52Hz |
| 23:30 | jnettlet | I am here. |
| 23:30 | jnettlet | pretty much what _rmk_ summed up. |
| 23:30 | otavio | :-) |
| 23:31 | _rmk_ | jnettlet: maybe we should create a community page on g+ for the carrier1? :) |
| 23:31 | dv_ | based on 2013.10 though, not 2013.07 (which is what is currently in OE master/dora) |
| 23:31 | dv_ | _rmk_: as said, some form of wiki or smth similar would be useful to keep track of these developments |
| 23:31 | jnettlet | a community page could be useful. |
| 23:32 | jnettlet | dv_, I can look if necessary but I don't think there are any major changes between 2013.07 and 2013.10 |
| 23:32 | _rmk_ | public? freely joinable? |
| 23:33 | _rmk_ | and I'm calling it "SolidRun Cubox-i Developments" |
| 23:33 | _rmk_ | any suggestions on changing anything? |
| 23:33 | jnettlet | sounds good to me. |
| 23:35 | jnettlet | dv_, actually there are a bunch of imx6 patches that are preferable to keep. Any imx6 build should really be using 2013.10 |
| 23:35 | dv_ | okay. so no isolating of patches and applying them on top of 2013.07 mainline. |
| 23:36 | dv_ | works for me |
| 23:38 | dv_ | alright, OE building started. this .. will take a while. |
| 23:38 | jnettlet | dv_, if you want I can consider the imx6 branch "stable" and create an imx6-devel branch for new stuff |
| 23:38 | dv_ | nah, thats fine |
| 23:39 | dv_ | I can specify which branch to check out anyway |
| 23:39 | dv_ | + a specific SRCREV |
| 23:39 | dv_ | I'm more curious about how well the patched linux-imx kernel works |
| 23:39 | dv_ | I will know in a few hours. |
| 23:40 | jnettlet | which reminds me. If anyone experiences any hangs etc with the latest u-boot let me know. I have re-implemented the clock gating settings from u-boot 2009. I am trying to figure out if they were the cause of some of my hangs or some other anomaly |
| 23:41 | _rmk_ | dv_: can you get to this? https://plus.google.com/communities/113793735711752568811 |
| 23:43 | _rmk | 23:43 * _rmk_ gives jnettlet two promotions :) |
| 23:43 | jnettlet | _rmk_, two wow? |
| 23:43 | _rmk_ | you may have a better photo for it - annoyingly g+ seems to have a desire for predominantly square photos |
| 23:44 | dv_ | joined |
| 23:44 | dv_ | and now I need to tend to other matters. I anyway cant do much while this is building. ttyl |
| 23:45 | _rmk_ | I'll share into it my posts so far |
| 23:46 | jnettlet | oddly enough I don't have a better photo. I think we can stick with yours. |
| 23:46 | jnettlet | Maybe rabeeh can get some PR type shots for us. |
| 23:49 | _rmk_ | yea, prefer something that doesn't involve my old worn out desk :) |
| 23:50 | _rmk_ | maybe this should be my next desk... https://fbcdn-sphotos-h-a.akamaihd.net/hphotos-ak-prn1/556660_527996387280857_1643533246_n.jpg :) |
| 23:51 | jnettlet | only if you use a touchpad. Glass desks and optical mice don't mix. |
| 23:52 | jnettle | 23:52 * jnettlet is going to retire to some bad television for a while. |
| 23:52 | jnettlet | ttyl |
| 23:52 | _rmk_ | true :) |