| 00:03.33 | ZaEarl | Look at what I found on my tablet motherboard: http://bit.ly/Lufbqa |
| 00:04.17 | hno | ZaEarl, Nice! |
| 00:05.01 | ZaEarl | that could be handy |
| 00:05.36 | hno | much |
| 00:06.18 | hno | which tablet is it? |
| 00:06.40 | ZaEarl | our ZaTab |
| 00:07.54 | ZaEarl | I'm tempted to prototype a case just for the motherboard. |
| 00:08.05 | ZaEarl | it's only 7cm x 10cm |
| 00:10.36 | hno | can it run without battery on usb power alone? |
| 00:10.57 | lundman | ok, lets attached the serial console |
| 00:10.59 | ZaEarl | I think I'm going to have to desolder a battery and find out |
| 00:22.05 | lundman | damnit cubie, this doesnt fit at all |
| 00:22.24 | hno | wrong cables? |
| 00:23.19 | lundman | the a2000 serial is sideways, not up. just a 6 lead cable with it, but the ends dont fit the pin thickness |
| 00:23.22 | lundman | I'll just use force |
| 00:26.17 | lundman | fits even worse on the board.. oh well |
| 00:30.04 | lundman | ahh ok, managed to convince it to play ball |
| 00:31.08 | hno | lundman, do you have the cable in the right way? The small ones should go to mele and big ones to the UART board.. |
| 00:31.55 | lundman | yep, done that now. i should be set |
| 00:32.13 | lundman | except the apple usb extender cable I found lying around, has a dimple in it, so usb dont fit |
| 00:32.19 | lundman | why is everything a hassle today |
| 00:33.38 | hno | Right.. those UART boards have a male connector on them... for plugging directly into an USB port. |
| 00:33.54 | hno | or hub- |
| 00:35.00 | hno | Night. |
| 00:35.27 | lundman | yay lights on |
| 00:35.30 | lundman | 115400? |
| 00:36.06 | hno | 115200 |
| 00:36.37 | *** join/#arm-netbook sepp (~seb@port-92-206-40-244.dynamic.qsc.de) |
| 00:37.02 | lundman | screen it is |
| 00:37.16 | hno | screen /dev/ttyUSB0 115200 |
| 00:37.57 | lundman | screen -S console -c /dev/null /dev/tty.usbserial 115200 |
| 00:37.59 | lundman | for me :) |
| 00:38.48 | lundman | great it works, didnt let me stop the boot though |
| 00:39.19 | lundman | ]wonder why its panicing |
| 00:39.32 | hno | what is panicing? |
| 00:40.52 | lundman | yeah its constantly panicing and rebooting.. but can't stop it :) |
| 00:41.06 | lundman | no key to stop uboot? |
| 00:42.10 | robws | lundman any key should stop it. are you sure keypresses are getting through ? |
| 00:42.30 | robws | mine was having that issue too |
| 00:42.53 | lundman | hard to know |
| 00:43.45 | lundman | stuck in the sd card to make it go different |
| 00:44.24 | lundman | [ 2.523220] Unable to handle kernel NULL pointer dereference at virtual address 00000019 |
| 00:44.27 | lundman | [ 2.541310] PC is at regulator_get_voltage+0x8/0x34 |
| 00:44.29 | lundman | oopsie |
| 00:45.04 | Turl | lundman: you using the qware leak as is? |
| 00:45.18 | lundman | no idea what that is |
| 00:45.19 | Turl | the axp20 regulator had some mismatching types that were fixed |
| 00:45.41 | Turl | where did you get your kernel from? |
| 00:45.49 | lundman | its ICS4 beta |
| 00:45.59 | lundman | i just assumed the pin outs on a2000 is same as a1000 |
| 00:46.01 | Turl | hm? |
| 00:46.03 | lundman | possible they are not |
| 00:46.26 | Turl | no idea about meles |
| 00:48.07 | lundman | the a1000 image is Gnd Rxt Txt +3v |
| 00:49.43 | libv | ZaEarl: cool! if that works, then the ZaTab and the mele become a good combination then for A10 display work |
| 00:50.22 | ZaEarl | libv, not sure why usb powered is interesting. There's an ac jack right next to it. Might as well use that. |
| 00:51.09 | libv | oh, this is for usb and not for uart? |
| 00:51.17 | killerkink | turl: where is the commit for the axp20 regulator? not on lichee-3.0.8-sun4i? |
| 00:51.45 | ZaEarl | I guess I need to verify that the motherboard works without a battery attached. whether that is powered by usb or powered by ac, that's secondary. |
| 00:52.04 | ZaEarl | and i need to make sure the uart ports work. |
| 00:52.50 | lundman | the pinouts are presumably the same, since the board is same revision |
| 00:54.17 | lundman | oh dont connect power heh |
| 00:59.31 | lundman | ok just took the optional metal plate of the extra hole at the back, so I can close the case again |
| 01:02.34 | Turl | killerkink: nope it's not there |
| 01:02.38 | lundman | no more panics, woo |
| 01:02.54 | Turl | https://github.com/amery/linux-allwinner/commit/24f8488f228a89b69579856d0328431487c446b3 |
| 01:04.47 | killerkink | turl: pardon my ignorance. what does it fixed in the axp? |
| 01:07.26 | killerkink | set to the lowerest voltage when suspend? |
| 01:09.55 | Turl | killerkink: the function type |
| 01:10.03 | Turl | compare the function definition before/after |
| 01:10.30 | killerkink | i am looking at it. :) |
| 01:13.06 | lundman | hey cool, my u-boot with ext4 and zfs appears to work |
| 01:13.17 | killerkink | the only thing that i see is that with the patch, its gurantee that suspend to choose the correct lowest voltage |
| 01:13.28 | lundman | U-Boot 2012.04.01-g39085db-dirty (May 29 2012 - 02:29:31) Allwinner Technology |
| 01:13.31 | lundman | zfsload - load binary file from a ZFS filesystem |
| 01:13.41 | killerkink | have the trace where the selector was set for other usecase |
| 01:14.13 | killerkink | apparently it not required to pass in the information in the axp20-reg.c |
| 01:31.41 | lundman | anyone have a uImage binary for me to boot |
| 01:50.26 | killerkink | turl: the selector is passed to the set voltage functions but not used? |
| 01:54.37 | Turl | killerkink: the patch just adjusts the function signature |
| 01:58.53 | killerkink | i see. but there is only one set_voltage? |
| 02:05.19 | *** join/#arm-netbook naobsd (6fbc2800@gateway/web/freenode/ip.111.188.40.0) |
| 02:14.55 | *** join/#arm-netbook hipboi (~hipboi@221.4.213.95) |
| 02:26.34 | lundman | sun4i#ext4load mmc 0:1 0x5000000 /boot/uImage |
| 02:26.34 | lundman | Loading file "/boot/uImage" from mmc device 0:1 (xxa1) |
| 02:26.36 | lundman | woo |
| 02:29.01 | lundman | sun4i#zfsload mmc 0:2 0x5000000 /boot/@/uImage |
| 02:29.02 | lundman | Loading file "/boot/@/uImage" from mmc device 0:2 xxa2 |
| 02:29.20 | lundman | great. now we can start working on it |
| 02:34.53 | steev | okay what's the best way to get the key codes for the vol -/+, power, back, home, and menu keys? trying getevent and it's just too chatty |
| 02:36.12 | Turl | steev: boot recovery |
| 02:36.19 | Turl | hit advanced menu |
| 02:36.20 | Turl | key test |
| 02:36.24 | Turl | press all the things! |
| 02:36.28 | Turl | write down the keycodes |
| 02:41.46 | steev | if it's "1" it's 001 or 100? |
| 02:42.26 | Turl | "1" is always "1" :P |
| 02:42.38 | lundman | or 31 |
| 02:42.51 | Turl | 1=0x1=0b1 |
| 02:43.14 | Turl | hno: wifi works OK |
| 02:43.30 | Turl | hno: it'll be good for mele + linux :) |
| 02:43.31 | steev | i'm trying to get this aroma installer working |
| 02:43.45 | steev | but it seems to be throwing a fit of i hate your touchscreen, even though you're using your updater-binary |
| 02:43.58 | Turl | hno: unfortunately it's not well suited to android |
| 02:44.11 | Turl | doesn't handle suspend nicely, the dongle just goes offline :P |
| 02:44.28 | Turl | steev: lol aroma installer |
| 02:45.08 | Turl | steev: I used it once, had to restore a stock(ish) ROM on my phone to test something |
| 02:45.22 | Turl | the touch was completely off |
| 02:45.30 | Turl | really crappy experience |
| 02:45.44 | steev | it works great on the gtab |
| 02:45.53 | Turl | besides the idea of having a wizard-y setup is just irritating |
| 02:46.17 | steev | well having a crappy vol - key, it makes it much easier to work with if touch worked |
| 02:47.04 | killerkink | steev: getevent <input device> |
| 02:48.01 | steev | killerkink: it's too chatty, what turl suggested is good |
| 02:48.05 | steev | i got the keys |
| 02:48.33 | killerkink | steev: not if you put in the <input device> that you want to monitor |
| 02:48.45 | *** join/#arm-netbook RITRedbeard (~RITReadbe@c-68-37-165-37.hsd1.nj.comcast.net) |
| 02:48.46 | steev | killerkink: i did put it in |
| 02:48.47 | killerkink | getevent /dev/input/event1 |
| 02:48.57 | steev | it just kept scrolling voer and over, unless i didn't want even3 |
| 02:49.02 | Turl | killerkink: still you get like 5-6 lines per touch |
| 02:49.27 | steev | going into recovery and doing key test gives you exactly what you look for |
| 02:49.27 | Turl | well gnite all :) |
| 02:49.53 | killerkink | hmm...yours is touch keys? |
| 02:50.06 | steev | nope |
| 02:50.29 | steev | i needed the key codes for vol -+, power, back menu home |
| 02:50.37 | steev | still don't have it working |
| 02:50.40 | steev | but meh |
| 02:52.06 | killerkink | if you its hard key, shouldnt there just be a press down/up event? |
| 02:52.37 | killerkink | just use the key in /system/usr/keylayout/sun4i-keyboard.kl |
| 02:53.17 | killerkink | if you have a working rom with the right keys |
| 03:17.34 | hipboi | Turl: have you compiled the xf86-video-mali ? |
| 03:17.41 | lundman | Concerns me a little, when the sata drivers has: for(tmp=0; tmp<0x3000; tmp++); |
| 03:59.14 | *** join/#arm-netbook hno (~hno@squid/developer/hno) |
| 04:31.46 | *** join/#arm-netbook CaCtus491 (~Kent@123-243-197-152.static.tpgi.com.au) |
| 04:52.02 | *** join/#arm-netbook CaCtus491 (~Kent@123-243-197-152.static.tpgi.com.au) |
| 05:46.17 | *** join/#arm-netbook gsilvis (~almostsix@50.12.163.241) |
| 05:59.26 | *** join/#arm-netbook Quarx (~Quarx@178.74.64.5) |
| 06:54.58 | *** join/#arm-netbook eebrah (~Ibrahim_K@41-139-199-130.safaricombusiness.co.ke) |
| 06:58.54 | hno | Turl, suspend/resume issues are fixable. |
| 07:00.03 | hno | and reloading your module hack should work around any issues there. |
| 07:16.03 | mnemoc | hno: thank you! |
| 07:16.40 | mnemoc | hipboi: afaik the userspace mali drivers need the DDK to be compiled |
| 07:17.02 | hipboi | mnemoc: yes |
| 07:17.04 | hipboi | i know |
| 07:17.08 | hipboi | i compiled ddk |
| 07:17.36 | hipboi | and the xserver-video-mali |
| 07:17.42 | hipboi | but when i run X |
| 07:17.47 | hipboi | i have some problem |
| 07:18.05 | hipboi | it looks like the mali drm driver can not find |
| 07:18.17 | hipboi | when i insmod mali_drm |
| 07:18.22 | hipboi | nothing happens |
| 07:18.38 | hipboi | no node is created in /dev |
| 07:20.12 | hipboi | http://paste.ubuntu.com/1028150/ |
| 07:20.40 | mnemoc | hipboi: did you enable DRM_MALI ? but as hno suggested yerterday our integration seems to be incomplete |
| 07:20.58 | hipboi | enable DRM_MALI ? |
| 07:21.05 | hipboi | just choose y in menuconfig |
| 07:21.18 | mnemoc | can you grep MALI .config ? |
| 07:21.49 | hipboi | CONFIG_DRM_MALI=m |
| 07:21.49 | hipboi | CONFIG_MALI=m |
| 07:22.13 | mnemoc | if you can share with use different userspace drivers compiled against r2p4-rel1 we can then focus in fixing the kernel integration in the open |
| 07:22.26 | mnemoc | s/use/us/ |
| 07:22.31 | hipboi | sure |
| 07:22.38 | mnemoc | excelent |
| 07:23.29 | specing | Have you checked what dmesg says? |
| 07:26.14 | hipboi | the problem is dmesg says nothing |
| 07:27.28 | specing | udev debugging? |
| 07:28.06 | mnemoc | I'm 99% sure it must be a problem in our naive drm_mali integration |
| 07:28.20 | specing | Heh ;) |
| 07:28.43 | hipboi | specing: how to use udev debugging |
| 07:30.26 | specing | hipboi: udevadm monitor |
| 07:30.40 | hipboi | i just downloaded the mali drm driver |
| 07:31.04 | hipboi | i found in the readme, the platform_device |
| 07:31.10 | hipboi | we did not add the platform_device |
| 07:31.26 | hipboi | : ) |
| 07:35.03 | *** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli) |
| 07:35.34 | hipboi | i am wrong |
| 07:35.45 | mnemoc | hipboi: we do, in sun4i's device.c |
| 07:36.00 | hipboi | yes, the platform_device is already there |
| 07:36.11 | mnemoc | hipboi: but doesn't connect this drm module with mali's memory space at all |
| 07:36.35 | mnemoc | it's like this drm_mali driver is merely a template |
| 07:36.50 | hipboi | yes |
| 07:36.55 | *** part/#arm-netbook hipboi (~hipboi@221.4.213.95) |
| 07:37.03 | *** join/#arm-netbook hipboi (~hipboi@221.4.213.95) |
| 07:37.25 | lundman | so its all going great |
| 07:38.28 | mnemoc | having android, armel and armhf user space drivers we can start improving the kernel integration |
| 07:38.46 | mnemoc | because more people will be able to work on it |
| 07:39.41 | hipboi | good |
| 07:41.25 | mnemoc | armhf cedarx drivers are also VERY welcomed :) |
| 07:41.37 | hipboi | yes |
| 07:41.42 | hipboi | i know this |
| 07:41.46 | hno | armhf is a requirement for any serious Linux usage imho. Running this device as armel is not fair to it's potentials. |
| 07:43.34 | hno | because any standard armel repository you find is built for armv5tel or similar cpu level, and you'll loose 30-50% of the userspace performance. |
| 07:44.21 | mnemoc | that's were r-pi becomes harmful |
| 07:44.32 | mnemoc | where* |
| 07:45.45 | mnemoc | those who can only maintain one arm repo they pick armel as it's "still relevant" (and on a hype) |
| 07:49.38 | hno | Well, those that really matters long-term on the Linux side of things focuses on armv7 hf and beyond. Not sure how things looks on the Android side. |
| 07:51.28 | mnemoc | no idea how dalvik deals with floats, but their native apps (NDK?) is probably stuck at armel |
| 07:52.29 | mnemoc | but it shouldn't really be a problem for the virtual machine to run on armhf |
| 07:53.08 | mnemoc | hopefully they will consider this transition for android 5.0 |
| 07:53.33 | mnemoc | as the hardware is been wasted |
| 07:55.25 | hno | I think Android is armel but optimized for armv7 as a compromise. Quite likely system components is even built targeting the target cpu in armel. This improves things considerably. |
| 07:56.26 | mnemoc | iirc there are quite a few armv5 and armv6 android devices |
| 07:56.38 | hno | In Linux it's different as many have taken the leap to hardfloat support for newer devices keeping armel for older CPUs. |
| 07:57.20 | hno | mnemoc, yes. armel built for armv5 but optimized for armv7. It's different dimensions in the CPU target selection to GCC. |
| 07:57.32 | mnemoc | i see |
| 07:59.05 | hno | also, you can run both at the same time, just not in the same process. |
| 07:59.19 | mnemoc | uh |
| 07:59.24 | mnemoc | that's fancy |
| 08:00.04 | hno | kernel do not care, but userspace ABI for libraries etc is incompatible |
| 08:01.08 | hno | i.e. you can run an armel chroot in an otherwise armhf setup. |
| 08:01.23 | mnemoc | so you mean multilib (/lib${abi}) |
| 08:02.06 | mnemoc | I thought there was magic in the binaries itself to make use of armv7 features when avaiable |
| 08:03.35 | hno | unfortunately there is no ABI flags in the binaries that multilib can easily use, so only works for chroots. |
| 08:04.20 | mnemoc | :( |
| 08:04.30 | hno | Yes it's a big :( |
| 08:05.01 | hno | What is even worse is that ld also do not notice, so you can link a binary with mixed code which will fail badly |
| 08:05.34 | mnemoc | ouch |
| 08:05.54 | mnemoc | I assumed this was an essential part of ELF |
| 08:06.21 | hno | It is, but ARM forgot... |
| 08:06.26 | *** join/#arm-netbook Mazon (~Mazon@95.166.104.217) |
| 08:06.39 | mnemoc | doh |
| 08:08.34 | specing | hno: armel? You mean armsf? |
| 08:09.04 | specing | ah, haven't scrolled down |
| 08:11.37 | hno | specing, easier to use armel / armhf names as those are more faimiliar. |
| 08:12.24 | lundman | doesn't seem likely the kernel would run any faster from armhf |
| 08:12.37 | mnemoc | kernel is float-free |
| 08:12.39 | specing | armel is arm little endian... |
| 08:12.46 | hno | and with ARMv8 there will be even more dimensions |
| 08:12.49 | lundman | yes |
| 08:12.58 | specing | armhf is arm hard float ABI |
| 08:13.28 | lundman | unless it refers to additional hardware math operations besides floats |
| 08:13.39 | hno | specing, I know. But armel also a repository type name that have been around forever in Linux world. |
| 08:13.47 | mnemoc | it's impossible to keep the sanity and been rigorus about the naming of arm optimizations |
| 08:14.04 | mnemoc | I prefer sanity |
| 08:14.15 | lundman | sanity is a luxury, but optional |
| 08:15.22 | hno | goes to work. |
| 08:15.32 | mnemoc | see you hno |
| 08:15.35 | lundman | so the "sata" drivers appears to be based on generic ahci work, with some tweaks |
| 08:15.52 | lundman | uboot ahci uses pci, so will have to do something different there |
| 08:18.15 | hno | lundman, there is much to generalise in uboot... the soc vendors tend to run a copy-pste party each time they add something to their target... |
| 08:18.25 | hno | s/pst/past/ |
| 08:19.09 | mnemoc | hno: the nand in u-boot seems much cleaner/saner than the one in linux.... can we use that to reimplement the linux part? |
| 08:19.36 | lundman | sure, I'm just looking around at the moment, getting familiar with it. |
| 08:20.04 | lundman | s/ /!/ |
| 08:20.10 | lundman | booo |
| 08:20.39 | lundman | s/o/O/g |
| 08:20.43 | lundman | yay |
| 08:21.23 | hno | misses regex support.. |
| 08:21.42 | hno | s/./!/g |
| 08:21.56 | hno | oops.. |
| 08:22.03 | mnemoc | hahaha |
| 08:35.49 | *** join/#arm-netbook drgreenthumb (~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net) |
| 08:38.57 | Gumboot | What kind of regex is that? The whole line should have come out as !!!. |
| 08:40.23 | mnemoc | seems ibot finds saner to assume the literal dot |
| 08:40.38 | mnemoc | s/\./!/g |
| 08:41.05 | mnemoc | ..x.. |
| 08:41.07 | mnemoc | s/\./!/g |
| 08:41.13 | mnemoc | s/./!/g |
| 08:41.45 | mnemoc | so there is no "any char" |
| 09:01.15 | hipboi | i found there is slightly difference between the mali_drm.c on github and from arm |
| 09:06.10 | mnemoc | same version? |
| 09:13.50 | *** join/#arm-netbook Bl1tter (~phys@48.Red-83-61-136.dynamicIP.rima-tde.net) |
| 09:44.42 | CIA-126 | rhombus-tech: Alban master * rb529c2c2f200 /allwinner_a10/orders/Alban.mdwn: |
| 09:48.38 | mnemoc | woh, a galician company interviewed in http://armdevices.net ! http://armdevices.net/2012/06/07/teofilio-romera-otero-of-igalia-open-source-software-company-at-computex-2012/ |
| 10:12.01 | *** join/#arm-netbook uwe_mobile (~uwe@static.88-198-8-117.clients.your-server.de) |
| 10:12.19 | *** join/#arm-netbook Jacmet (~peko@stolen.plutonium.dk) |
| 11:40.00 | *** join/#arm-netbook tom_say (~quassel@cpe-68-203-248-184.stx.res.rr.com) |
| 11:41.39 | *** join/#arm-netbook polvo_ (~chatzilla@94.177.23.95.dynamic.jazztel.es) |
| 12:39.32 | *** part/#arm-netbook Bl1tter (~phys@48.Red-83-61-136.dynamicIP.rima-tde.net) |
| 12:55.33 | lkcl | ManoftheSea: it's a long story. that was the first project where i learned that respected prominent free software project leaders can also be bullies. |
| 12:59.47 | *** join/#arm-netbook Guest91346 (~mathew@cpc8-rdng20-2-0-cust313.15-3.cable.virginmedia.com) |
| 12:59.53 | Guest91346 | hi all |
| 13:02.18 | dotmat | Im having a major issues trying to format an SD Card to use within an AllWinner A10 ARM chipset. Does anyone know of a straight forward guide or has 10 mins guide me though the process? |
| 13:18.57 | *** join/#arm-netbook killerkink (af9cdba0@gateway/web/freenode/ip.175.156.219.160) |
| 13:28.21 | specing | dotmat: 1) fire up fdisk 2) format 3) ... 4) profit! |
| 13:29.41 | mnemoc | dotmat: define "major issues" |
| 13:30.47 | dotmat | having never used fdisk or dd im being a bit cauations but I cant seem to get the partitions set up right |
| 13:31.28 | dotmat | ive compiled the OS, but it seems Im not setting the right partition sizes and setting uboot correctly |
| 13:31.28 | specing | dotmat: there is gparted for the fainthearted |
| 13:31.34 | specing | nice little gui |
| 13:53.43 | *** join/#arm-netbook hp_ (~kvirc@p5B389FA3.dip.t-dialin.net) |
| 14:08.57 | *** join/#arm-netbook mathew (~mathew@cpc8-rdng20-2-0-cust313.15-3.cable.virginmedia.com) |
| 14:28.49 | *** join/#arm-netbook hark (~hark@w.v89.eu) |
| 15:34.43 | *** join/#arm-netbook QingPei (~qingpei@123.118.110.183) |
| 16:03.38 | *** join/#arm-netbook naobsd (7c294571@gateway/web/freenode/ip.124.41.69.113) |
| 16:35.35 | *** join/#arm-netbook mathew (~mathew@cpc8-rdng20-2-0-cust313.15-3.cable.virginmedia.com) |
| 17:58.23 | *** join/#arm-netbook mathew (~mathew@cpc8-rdng20-2-0-cust313.15-3.cable.virginmedia.com) |
| 18:19.46 | *** join/#arm-netbook Tsvetan (5e488b19@gateway/web/freenode/ip.94.72.139.25) |
| 18:20.00 | Tsvetan | hi |
| 18:20.48 | mnemoc | hi Tsvetan |
| 18:33.14 | *** part/#arm-netbook QingPei (~qingpei@123.118.110.183) |
| 18:33.50 | hno | hi Tsvetan |
| 19:00.50 | *** join/#arm-netbook rell01 (~chatzilla@p4FE56247.dip0.t-ipconnect.de) |
| 19:05.11 | steev | woo getting closer |
| 19:05.26 | steev | not quite sure what's causing my lockup yet |
| 19:05.37 | steev | but it boots and loads the modules with 3.0.31 |
| 19:05.44 | steev | i do have one patch to submit |
| 19:05.56 | steev | mma7660 requires INPUT_POLLDEV |
| 19:08.47 | Turl | steev: nice to know :) |
| 19:09.23 | Turl | does the mele have wifi or just ethernet? |
| 19:10.01 | steev | Turl: well it's exciting to use "my own" rom |
| 19:10.11 | steev | based on you and ZaEarl's work |
| 19:10.37 | steev | and it gives me something to work on while i do this stupid devicetree stuff for work |
| 19:10.43 | steev | devicetree is stupid. |
| 19:11.08 | Turl | are you using the common repo? |
| 19:11.24 | steev | yeah |
| 19:11.29 | Turl | good |
| 19:11.38 | steev | my a710 directory is a bit of a rip from the zatab one |
| 19:11.51 | steev | and then i made an a710 defconfig since i use the mma7660 instead of bma250 |
| 19:11.57 | Turl | yeah that's the intention of making a common repo :P |
| 19:13.48 | steev | bah |
| 19:14.12 | steev | why doesn't android handle module dependencies, so stupid |
| 19:14.31 | Turl | steev: it's not android's fault |
| 19:14.38 | Turl | steev: insmod on your PC doesn't handle them either |
| 19:14.39 | Turl | :P |
| 19:14.40 | steev | uh |
| 19:14.50 | steev | right, but why does it use insmod instead of modprobe |
| 19:15.05 | Turl | you can use modprobe I guess |
| 19:15.09 | Turl | but I think it's busybox provided |
| 19:16.49 | steev | yeah, i'll just change the defconfig and make it a Y for INPUT_POLLED |
| 19:36.49 | *** join/#arm-netbook naGt (~nacho@151.225.218.87.dynamic.jazztel.es) |
| 19:37.02 | naGt | hi all! |
| 19:37.58 | naGt | I've just booted my first debian in mele hardware, but any time I try to link up eth0, I get a "link not ready" message. Is this normal? Is hardware supported? |
| 19:40.22 | mnemoc | naGt: be sure to be using the stock script.bin, the one in Tom's ubuntu image doesn't have ethernet support |
| 19:40.58 | naGt | mnemoc, It has much sense, because I'm re-using the Tom's script.bin |
| 19:41.20 | dotmat | does the sun4i repository still exist on github? I cant seem to find it - the only one i can see is murphy666 / sun4i-kernel. |
| 19:41.48 | mnemoc | dotmat: https://github.com/amery/linux-allwinner ? |
| 19:41.56 | naGt | mnemoc, The stock script.bin is in the nand, isn't it? |
| 19:42.04 | mnemoc | nanda, yes |
| 19:42.27 | mnemoc | naGt: or https://github.com/amery/sunxi-bin-archive/tree/master/mele-a1000/stock-nanda |
| 19:42.47 | mnemoc | but you may need to rename it to what your bootloader expects |
| 19:43.03 | mnemoc | the uboot in Tom's image wanted evb.bin |
| 19:43.11 | dotmat | mnemoc - im using the guides posted on rhombus-tech are those out of date? |
| 19:43.30 | mnemoc | dotmat: I don't know... url? |
| 19:43.41 | dotmat | http://rhombus-tech.net/allwinner_a10/hacking_the_mele_a1000/Building_Debian_From_Source_Code_for_Mele/ |
| 19:43.55 | naGt | mnemoc, Ok, I forgot the "nanda" name, despite it's easy to remember |
| 19:44.24 | mnemoc | nand a,b,c,d,e,... :) |
| 19:45.15 | mnemoc | dotmat: the kernel part looks ok. `git checkout allwinner-v3.0-android-v2` is not needed as it's the default branch |
| 19:46.15 | mnemoc | personally I prefer linaro's toolchain, but shouldn't affect |
| 19:46.26 | mnemoc | it can be 4.6 and even 4.7 |
| 19:46.39 | mnemoc | no idea why that guide suggests 4.4 |
| 19:47.44 | *** join/#arm-netbook jimyx17 (~jaime@139.124.223.87.dynamic.jazztel.es) |
| 19:52.22 | mnemoc | dotmat: what problem did you have with https://github.com/amery/linux-allwinner ? |
| 19:53.21 | dotmat | mnemoc, ive not had any prob with it, all the documents ive been using refer to 'git checkout sun4i' |
| 19:53.37 | mnemoc | dotmat: that's the uboot's branch name |
| 19:54.07 | mnemoc | things in the kernel tree are sadly much more complicated branch-wise |
| 19:54.21 | dotmat | ok, thats still the bit im trying to get my head around. |
| 19:54.43 | mnemoc | dotmat: but the wiki you refered to had the right 'checkout' command |
| 19:57.07 | dotmat | when i type git checkout sun4i into terminal it returns 'Already on 'Sun4i' |
| 19:57.09 | naGt | mnemoc, :D, now I have ethernet connection, thanks a lot again!! |
| 19:57.25 | mnemoc | naGt: yw :) |
| 19:57.52 | naGt | mnemoc, now the next target if possible, wireless connection |
| 19:58.20 | mnemoc | dotmat: git checkout is to jump from one branch to another. if you jump to the current you'll see that message |
| 19:58.26 | Turl | mnemoc: btw, you can use the mainline rtl wifi driver with a little tweaking :) |
| 19:58.28 | mnemoc | dotmat: there is nothing wrong about that |
| 19:58.33 | mnemoc | Turl: -v |
| 19:59.55 | dotmat | Turl command not found - do I need to install another package? |
| 20:00.23 | Turl | dotmat: hm? |
| 20:00.53 | dotmat | sorry i think im getting peoples names and terminal commands mixed up. |
| 20:01.39 | Turl | lol |
| 20:01.57 | dotmat | this is a bit more complex than my arduino lol |
| 20:02.32 | Turl | that reminds me, I need to call the post and figure out where's my arduino ~.~ |
| 20:04.10 | steev | wish i knew where my tablet was |
| 20:06.11 | steev | the "tracking number" it lists is the paypal transaction id :/ |
| 20:06.18 | Turl | lol |
| 20:06.31 | mnemoc | Turl: what tweaking? |
| 20:06.42 | steev | but it was shipped on the 5th so i'm assuming it will be here next week some time |
| 20:07.27 | Turl | mnemoc: https://github.com/allwinner-dev-team/linux-allwinner/commit/47e120c7ab0957ce608f90aaeae4236d6868745c |
| 20:08.16 | mnemoc | wtf |
| 20:09.39 | Turl | after that the device magically gets powered, connects and the mainline driver picks it up, works perfectly :) |
| 20:11.48 | Turl | mnemoc: well you also need the firmware on your filesystem, but that's rather trivial to set up |
| 20:11.57 | Turl | might even be there already if you're using a distro |
| 20:12.48 | mnemoc | so the chip is connected to a usb controller which is disabled by default? |
| 20:13.50 | mnemoc | and all the other ports come from the same hub? |
| 20:14.15 | Turl | yeah, the chip is connected over USB |
| 20:14.32 | Turl | can't comment on the other ports, haven't looked into it much |
| 20:14.37 | mnemoc | your hack doesn't do anything rtl8192cu specific |
| 20:14.56 | mnemoc | it simply queries script.bin and enables a usb controller |
| 20:14.58 | Turl | mnemoc: yeah, but I copied the magic call from the rtl8192cu driver |
| 20:15.11 | Turl | no idea if other drivers are connected over usb |
| 20:15.17 | Turl | s/drivers/devices/ |
| 20:15.45 | mnemoc | I have to move that generic voodoo to mach-sun4i/devices |
| 20:16.32 | mnemoc | if the reason the rtl8192cu is not see is because the usbc is disabled nothing of what works can be hanging from there |
| 20:17.01 | mnemoc | seen* |
| 20:18.28 | Turl | mnemoc: from what I gather using lsusb, there is like 5 usb buses on the tablet |
| 20:18.49 | Turl | bus 2 has a 1a40:0101 hub too apparently |
| 20:18.56 | Turl | the rtl is on bus 4 |
| 20:19.01 | mnemoc | is rtl8192cu alone? |
| 20:19.45 | Turl | yeah, there's just a 1d6b:0002 Linux Foundation 2.0 root hub too, but pretty sure that's a dummy device :) |
| 20:20.07 | Turl | $ lsusb|grep 004 |
| 20:20.08 | Turl | Bus 004 Device 001: ID 1d6b:0002 |
| 20:20.08 | Turl | Bus 004 Device 002: ID 0bda:8176 |
| 20:21.07 | Turl | mnemoc: script.fex has the usb wifi entry |
| 20:21.09 | Turl | [usb_wifi_para] |
| 20:21.09 | Turl | usb_wifi_used= 1 |
| 20:21.09 | Turl | usb_wifi_usbc_num= 2 |
| 20:21.26 | Turl | there's a sdio section too, but that's used=0 here |
| 20:21.38 | Turl | I suppose other tablets could have sdio based wifi chips |
| 20:23.28 | mnemoc | very interesting |
| 20:26.15 | mnemoc | should remove that tab from bin2fex |
| 20:30.21 | *** join/#arm-netbook EricBetts (bettse@gateway/shell/sudoers/x-fsyertesjctmvhms) |
| 20:33.38 | *** join/#arm-netbook frantisheq (~androirc@kapusany.3log.sk) |
| 20:56.21 | steev | wish the tablet would get here so i can take this a710 apart |
| 20:56.47 | steev | i want to know why it's locking up |
| 21:09.20 | rz2k | locking up = freeze with only power cycle option left? |
| 21:10.02 | rz2k | my ly-f1 seems to have something like that, I thought it was due to crappy flex pcb connector to LCD |
| 21:10.21 | mnemoc | it can also be that you get trapped in FEL |
| 21:12.52 | rz2k | also it is completely random, sometimes tablet works for hours in android or my testing ubuntu, sometimes locks up on uboot(?) bootscreen or even before that (on poweron I see short LCD backlight flash, in 1-2 seconds bootscreens start to appear). |
| 21:14.52 | rz2k | weirdest thing was when it died on flashing process and I was unable to flash new rom without crossing DQ lines on NAND. |
| 21:15.53 | mnemoc | ouch |
| 21:24.20 | hno | Turl, your module hack enables the A10 USB controller and USB power. |
| 21:24.58 | hno | there is three controllers, one of which is OTG capable. |
| 21:28.34 | Turl | hno: nice to know |
| 21:29.28 | hno | I would not expect there to be an usb hub in a tablet. |
| 21:30.02 | hno | there is one in the mele however, to provide 3 USB host ports. |
| 21:30.54 | hno | actually it's 4 USB HOST ports + wifi + OTG., |
| 21:31.20 | mnemoc | tablet might use the same "reference design" |
| 21:31.31 | mnemoc | hub + wifi + otg |
| 21:32.10 | hno | i doubt. usb is very easy. |
| 21:32.45 | hno | and the hub chip costs a bit. |
| 21:33.29 | mnemoc | Turl: beside the wifi, the generics and what you connect, what else lsusb shows? |
| 21:34.07 | hno | there is sufficient interal usb ports for wifi + otg + host port. |
| 21:34.09 | Turl | a hub |
| 21:34.46 | Turl | terminus technology inc |
| 21:35.09 | mnemoc | http://www.terminus-tech.com/English/company.html |
| 21:35.17 | Turl | # adb shell lsusb|grep -v 1d6b |
| 21:35.17 | Turl | Bus 002 Device 002: ID 1a40:0101 |
| 21:35.20 | Turl | this is wifi off |
| 21:35.39 | Turl | # adb shell lsusb|grep -v 1d6b |
| 21:35.39 | Turl | Bus 002 Device 002: ID 1a40:0101 |
| 21:35.40 | Turl | Bus 004 Device 002: ID 0bda:8176 |
| 21:35.40 | Turl | wifi on |
| 21:35.41 | rz2k | ly-f1 internal usb schematic http://sdl4.4pda.ru/1755842/image/jpeg/OV.JPG?000268965f199ef64fd11e6e00000000c9658ea0dfb4f4979d5de853db845aea |
| 21:35.43 | mnemoc | so they have a hub connected to nothing??? |
| 21:38.07 | hno | mele says |
| 21:38.09 | hno | Bus 001 Device 001: ID 1d6b:0002 |
| 21:38.09 | hno | Bus 002 Device 001: ID 1d6b:0001 |
| 21:38.09 | hno | Bus 001 Device 002: ID 05e3:0608 |
| 21:38.09 | hno | Bus 003 Device 001: ID 1d6b:0002 |
| 21:38.09 | hno | Bus 003 Device 002: ID 0bda:8176 |
| 21:38.10 | hno | Bus 004 Device 001: ID 1d6b:0001 |
| 21:38.36 | hno | and thats with only wifi on the usb bus |
| 21:38.44 | hno | + a 4 port hub, |
| 21:38.46 | Turl | mnemoc: the otg is the hub apparently |
| 21:38.49 | Turl | # adb shell lsusb|grep -v 1d6b |
| 21:38.49 | Turl | Bus 002 Device 002: ID 1a40:0101 |
| 21:38.50 | Turl | Bus 002 Device 004: ID 22b8:2d66 |
| 21:38.55 | Turl | that's with my phone on OTG |
| 21:38.58 | Turl | and wifi off |
| 21:40.20 | hno | Turl, phone or tablet as otg master? |
| 21:40.35 | Turl | tablet |
| 21:40.38 | Turl | phone can't do otg |
| 21:40.47 | hno | guessed so |
| 21:41.16 | Turl | 22b8:* = motorola |
| 21:43.37 | DonkeyHotei | 1a40:0101 is a 4-port hub |
| 21:44.10 | DonkeyHotei | so is 05e3:0608 |
| 21:44.11 | mnemoc | connected to nothing :-\ |
| 21:44.36 | Turl | http://usb-ids.gowdy.us/read/UD/1a40/0101 |
| 21:44.43 | mnemoc | well... the mele has an unused ethernet chip |
| 21:44.50 | hno | DonkeyHotei, the mele have a 4 port hub. |
| 21:45.02 | hno | mnemoc, what? |
| 21:45.06 | DonkeyHotei | i doubt the ethernet would be on usb |
| 21:45.18 | Turl | DonkeyHotei: raspberry pi's ethernet is over usb |
| 21:45.23 | DonkeyHotei | ugh |
| 21:45.37 | mnemoc | hno: the realtek next to the chip is not used |
| 21:45.56 | mnemoc | hno: we get ethernet using the wemac thing within the soc |
| 21:45.56 | hno | mnemoc, it's the phy for the ethernet port. |
| 21:45.57 | DonkeyHotei | what realtek, and what ethernet? |
| 21:46.17 | mnemoc | hno: uhm... |
| 21:46.41 | mnemoc | DonkeyHotei: rtl8201cp |
| 21:46.45 | Turl | mnemoc: zatab has two identical gt801 goodix chips |
| 21:47.32 | hno | A10 only have a MAC. The PHY is needed for TP connection, |
| 21:47.34 | DonkeyHotei | rtl8201cp is a phy |
| 21:48.02 | mnemoc | i see... so they aren't stupid after all |
| 21:48.14 | mnemoc | :) |
| 21:49.27 | hno | have not found anything with more than two legs unused in the mele, not counting those three internal connectors and ESD protection for those. |
| 21:50.20 | hno | stil a mystery what that 6 pin internal usb header is meant to be used for. |
| 21:51.00 | DonkeyHotei | usb headers would presumably be for usb |
| 21:52.04 | hno | yes, but why 6 pins? d+/d-/gnd/vusb/??/?? (it's a host port, so no id pin) |
| 21:53.15 | DonkeyHotei | are you sure they're connected? |
| 21:53.58 | hno | yes. one is probably v3_3 |
| 21:54.26 | DonkeyHotei | and an extra ground? |
| 21:54.52 | hno | not sure, have not spent too much time on it. |
| 21:55.17 | hno | could be |
| 22:29.50 | *** join/#arm-netbook frantisheq (~androirc@kapusany.3log.sk) |
| 22:38.01 | *** join/#arm-netbook ZaEarl_ (~malmrose@173-11-125-162-SFBA.hfc.comcastbusiness.net) |
| 22:53.46 | A1Sheds | http://armdevices.net/2012/06/07/windows-8-rt-on-a-68-7-via-8850-arm-cortex-a9-tablet-by-eken/ |
| 23:01.00 | ZaEarl_ | lol, win8 theme for android |
| 23:01.45 | DonkeyHotei | i hope MS sues for trademark infringement |
| 23:02.00 | ZaEarl_ | sue a chinese company? |
| 23:02.19 | DonkeyHotei | sue any US vendor with it |
| 23:02.33 | libv | this one is better: http://www.engadget.com/2012/06/07/thd-n2-kirf-macbook-air-ics/#continued |
| 23:02.41 | libv | go on apple! sue them! |
| 23:02.57 | libv | it actually does have an A10 |
| 23:20.16 | A1Sheds | $44 capacitive Rk2906 ICS tablet ! |
| 23:20.25 | A1Sheds | http://armdevices.net/2012/06/07/44-capacitive-rk2906-ics-tablet-by-sawink-at-computex-2012/ |
| 23:20.38 | *** join/#arm-netbook PhantomsDad (~PhantomsD@c-69-255-33-53.hsd1.md.comcast.net) |
| 23:20.39 | *** join/#arm-netbook PhantomsDad (~PhantomsD@konversation/developer/PhantomsDad) |
| 23:21.04 | A1Sheds | It looks like the 1K pieces pricing but they are getting lower |
| 23:23.04 | A1Sheds | libv: I'd like to try one of the A10 $149 notebooks, it might have a decent display and case |
| 23:32.45 | libv | HDReady display might not be know to the kernels we have, i haven't looked |
| 23:33.22 | libv | known even |
| 23:35.43 | A1Sheds | it looks like a fun few years ahead |
| 23:36.15 | *** join/#arm-netbook ZaEarl_ (~malmrose@173-11-125-162-SFBA.hfc.comcastbusiness.net) |
